10 Commits

Author	SHA1	Message	Date
David Blass	4d1fd5ea1a	fix: 4 unaddressed log-audit / run-audit findings + close 10 already-resolved issues (#785) ... * fix: 4 unaddressed log-audit / run-audit findings closes 4 issues with code changes; 7 issues are already addressed by #769 and 3 are deferred — see PR description. #782 Anthropic 401 → `isApiKeyAuthError` now matches the direct-Anthropic 401 shape (`Failed to authenticate. API Error: 401 ...`, `authentication_error`, `Invalid bearer token`, `api_error_status=401`) so revoked / mistyped / rotated `ANTHROPIC_API_KEY` users see the formatted rotate-key CTA instead of a raw 401 JSON dump. #778 billing-class provider errors → `providerErrors.ts` now classifies `CreditsError` / `FreeUsageLimitError` / `Insufficient balance` / `spending cap` as `provider billing exhausted` *before* status-code patterns can win and tag them as transient `auth error (401)` / `rate limited (429)`. `agentHangReport.ts` swaps the bare "Pullfrog stalled — auth error" headline for a billing-specific CTA (extracts the provider's billing URL when present). #775 silent IncrementalReview swallows `BillingError` → `reportErrorToComment` now optionally falls through to creating a fresh issue comment on `toolState.issueNumber` when no progress comment exists. Wired with `createIfMissing: true` from the `BillingError` / `TransientError` paths in `proxy.ts` so silent triggers (`pull_request_synchronize`) finally surface the router-balance signal on the PR instead of only in the GH job summary. #773 `currentUser()` inside `after()` → `fillInstallerIdentityIfMissing` is split into `resolveInstallerIdentity` (must run inside the request body) and `fillInstallerIdentity` (DB-only, safe in `after()`). The `/console/[owner]` caller now resolves Clerk identity up-front and defers only the prisma write, fixing the broken installer-identity backfill on org-console first-admin visits. Co-authored-by: Cursor <cursoragent@cursor.com> * add /audits cursor command for triaging run-audit + log-audit issues Co-authored-by: Cursor <cursoragent@cursor.com> * review prompt: tighten body-section bar + inline technical-details (#770) * review prompt: tighten body-section bar + add inline technical-details Two layers of tightening to the Review/IncrementalReview prompts in PR_SUMMARY_FORMAT (and the per-mode aggregate-&-draft step): 1. Reframe inline-vs-body split. Body `### ` sections are now reserved for concerns that genuinely have no line to anchor to — absence, sequencing, design decisions, scope questions, architectural risk. Drop the "cross-cutting concerns" framing (misled the agent into either filing nothing in the body or filing multi-file anchored findings there). 2. Add a "Hunt for non-anchored concerns" sub-step to both Review (step 6) and IncrementalReview (step 8) aggregate phases. Diagnosis from PR #767's auto-review: on substantial PRs the agent surfaced findings but routed all of them inline, producing reviews with zero `### ` body sections even on diffs where non-anchored concerns clearly existed. 3. Replace the abstract `### ` example with a concrete non-anchored one ("Legacy `opencode.ts` has no documented deletion plan") so the agent pattern-matches the absence-shaped finding, not a line-bug. 4. Add an "Inline technical details" subsection to PR_SUMMARY_FORMAT so inline comments can carry a `<details>Technical details</details>` block when the fix has cross-file implications. Rename the existing "Agent details" inline collapsible to "Technical details" for consistency with body sections. 5. (Carried over from prior uncommitted work) Restructure the review metadata block from `<details>Review metadata</details>` into an HTML comment + an italic TL;DR commit-range line. The HTML comment keeps the metadata addressable for downstream agents without eating user-visible review real estate. No tests touched. * wiki: document multi-model end-to-end eval pattern * feat(promo): cookie-stashed promo codes for onboarding rewards (#771) * feat(promo): cookie-stashed promo codes for onboarding rewards Operator hands out a link like https://pullfrog.com/start?promo=FROGGY; middleware validates the code against an in-code registry, stashes it in an HttpOnly cookie, and the install callback applies the reward once the GH-side account exists. v1 reward: unlimited_runs (lifts the monthly free-runs cap to 1M, same convention prod-grandfathered accounts use). No schema changes. Idempotent across reinstalls via the lte: 100 gate. * fix(promo): integrate handler into existing proxy.ts (Next 16 rename) * docs(promo): clarify sentinel + sync plan doc with renamed paths * feat(promo): add FOUNDATIONS code * feat(promo): show applied promo code in console * refactor(promo): move cookie set to client-side * docs(promo): point JSDocs at PromoCookieSetter, not proxy.ts * billing: cap counts only successful runs (#787) * billing: cap counts only successful runs `reserveRun` was counting `WorkflowRun` rows regardless of status against `Account.includedMonthlyRuns`. Failed / cancelled / skipped / timed-out runs consumed cap slots even though their `billableCents` got zeroed on the completion webhook — pushing paying users into billable territory earlier than the contract implies. `inthhq` paid for 2 extra runs this month because 2 failed runs ate 2 of their 100 free slots. Cap query now filters on `CAP_CONSUMING_STATUS = "success"`. Only runs that actually deliver value consume slots; in-flight (`running`) runs hold no slot until they terminate as success (burst-bypass risk is theoretical given GH Actions concurrency limits). Shared constant lives in `utils/billing.ts` and is used in lockstep by three call sites: `reserveRun` (live cap gate), the billing API's `runsThisMonth` (dashboard progress bar), and the billing-report script's `cap` column. Script's `cap` cell was also broken independently — it compared `monthBillableRuns` (overage count) against `includedMonthlyRuns` (free cap), so `inthhq` rendered as `125/100 (over)` when the meaningful ratio is `223/100 (over)`. Fixed to use `mRuns/cap`, which is the same predicate the live billing path uses. * move CAP_CONSUMING_STATUS to workflowRunStatus.ts + wire script through it Per copilot review: the JSDoc claimed the billing-report script used the constant in lockstep, but the script kept `status: "success"` inline. The script imports from raw-node ESM and can't pull in `next/server`, so it couldn't import from `utils/billing.ts`. Moved the constant to `utils/workflowRunStatus.ts` (already Next-free, already the home of `CONCLUSION_VALUES`) and updated all three call sites to import from there. Script's `mRuns` query now uses `CAP_CONSUMING_STATUS` directly, making drift impossible. * learnings: audit fixes — preamble in TOC, server-side line-boundary truncation, empty-repo intro (#743) * learnings: surface preamble in TOC, mirror line-boundary truncation server-side, fix empty-repo intro copy three audit fixes on top of the recent learnings overhaul (#717): - `parseLearningsHeadings` now prepends a synthetic `(preamble)` entry when a body has non-whitespace content before the first heading. the prompt instructs the agent NOT to slurp the whole file when a TOC is present, so without this any preamble lines were silently invisible (realistic transitional case: an agent partially restructures a legacy free-text body and leaves bullets above the first `## `). - server-side PATCH route now applies the same line-boundary-aware truncation as the action (defense in depth via a shared `truncateAtLineBoundary` + `MAX_LEARNINGS_LENGTH` exported from `action/internal`). the raw `.slice` it used before could leave a mid-heading tail on any caller that bypassed the client-side truncate, breaking the next-seed TOC parse. removes the duplicated cap constant. - `buildLearningsSection` intro no longer asserts "accumulated by previous agent runs" — false for fresh repos with zero history. new copy is tense-neutral and works for empty + populated bodies. also nudges the agent to re-read after mid-run edits (the inlined TOC ranges are a run-start snapshot). Co-authored-by: Cursor <cursoragent@cursor.com> * learnings prompt: tighten to single evergreen test, allow tool-quirk bullets when they prevent repeat waste The blanket "no pullfrog tool quirks" ban was wrong — if the agent burned calls discovering a quirk this run, recording the workaround prevents the next run from repeating the waste. Reframe around one litmus ("would a future run do its work better because this bullet exists?") and trust it to subsume the scattered don'ts. Drop the 3+ months timeframe (arbitrary) and the four-example pullfrog/PR/date/play-by-play list (the rule underneath is "don't anchor facts to repo state that will move"). Cuts ~10 lines from a prompt the model was already mostly ignoring; the remaining anchor list is narrower and more enforceable. * audit-learnings-r2: align wiki + tighten re-read nudge - wiki/prompt.md described the post-run reflection prompt as "bans pullfrog-tool quirks (those belong in tool descriptions, not per-repo learnings), bans PR/review/commit/date references" — that's stale after the prompt rewrite. update to: single-litmus framing, expanded anchor list (now includes version pins + line numbers), and explicit allowance for tool-quirk workarounds when discovery burned calls. - buildLearningsSection re-read nudge said "re-read after editing" which can be read as "re-read the section you edited". in fact any edit shifts the line numbers of every later section in the TOC, not just the edited one. tighten to make that explicit. mirror the new wording in the wiki example block. update the test substring assertion accordingly. * postRun: refresh JSDoc to match the reflection prompt rewrite `buildLearningsReflectionPrompt`'s JSDoc still listed "PR-/review-/commit-/date-anchored facts" and "rediscovery of pullfrog-tool quirks" as failure modes the prompt pushes back on. after b586b4f8 the prompt no longer bans tool-quirk bullets (it explicitly allows them when the agent burned calls discovering the quirk), and the anchor list expanded to cover branch refs, version pins, and line numbers too. update the JSDoc so it describes the prompt that actually exists, and call out the cross-repo drift tradeoff that comes with allowing tool-quirk bullets. * fix(mcp/issueEvents): narrow event.event before Set.has lookup octokit's listEventsForTimeline union includes timeline-event members where `event` is `event?: string`. `("event" in event)` does not narrow that property to non-undefined, so `relevantEventTypes.has(event.event)` was passing `string | undefined` to a `Set<string>.has`. typescript only flagged this once `cf-worker-indexing` started seeing the file via the type graph that now reaches mcp through the new `truncateAtLineBoundary` re-export in `action/internal/index.ts`. fix the latent bug at the source: require `typeof event.event === "string"` before the Set lookup. * learnings: split truncation helpers into MCP-free module re-exporting `truncateAtLineBoundary` + `MAX_LEARNINGS_LENGTH` from `action/utils/learnings.ts` through `action/internal/index.ts` accidentally pulled the entire MCP type graph into the SDK barrel: `learnings.ts` imports `ToolContext` from `mcp/server.ts`, which transitively wires every tool module under `action/mcp/` into anything that imports from `pullfrog/internal`. for `cf-worker-indexing/tsconfig.json` (`customConditions: ["@pullfrog/source"]`) and the root `tsc` (which compiles the proprietary app routes that import from `pullfrog/internal`), this expanded the type-checked surface and surfaced two latent issues in unrelated files (`mcp/issueEvents.ts`, `utils/subprocess.ts`). a 6-line pure string helper has no business dragging mcp/server.ts into anyone else's type graph. move both symbols to `action/utils/learningsTruncate.ts`. `learnings.ts` re-exports them so existing callers keep working; `internal/index.ts` re-exports from the truncate-only module so the SDK barrel stays MCP-free. --------- Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Colin McDonnell <colinmcd94@gmail.com> * trim first-run celebration email to short personal note drops the feature-dump bullet list (custom review instructions, github iteration walkthrough, security model) — wrong moment to teach. keeps the congrats, the reply CTA, adds discord/x links, keeps the router credit P.S. handler no longer needs the workflowRun→repo lookup. * signup-report: per-bucket histogram Adds a UTC-aligned signups-per-bucket histogram between the overview block and the company-email list. Empty buckets are pre-filled with 0 so dry spells render as gaps. New `BUCKET=hour|day` env flag with a smart default (hour if window ≤ 48h, else day). Histogram is also included in the JSON payload under `histogram: [{key, count}, ...]`. * signup-report: drop hourly bucket, day-only histogram * feat(billing): monthly Router spend limits (hard-cap + alert-only) (#748) * feat(billing): monthly router spend limits (hard-cap + alert-only modes) (#660) Per-account ceiling on the sum of `router_topup` invoices (pending + succeeded) for the current UTC calendar month. Closes a gap where a runaway agent loop, leaked PR trigger, or stuck workflow could auto-reload indefinitely with no aggregate per-month ceiling. Two enforcement modes via `RouterLimitMode` enum: - `hard_cap`: refuse new auto-reloads; PR comment via reserveRun; 402 `router_monthly_limit` from /api/proxy-token; email + banner - `alert_only`: auto-reload keeps flowing; email + banner only, first breach per UTC month Enforcement is split across reserveRun (pre-dispatch paywall comment) and /api/proxy-token phase-1 (mid-run 402). Both surfaces read through the same `getRouterSpentThisMonthCents` helper so the dashboard, the dispatch gate, and the auto-reload gate can't disagree. Email dedup uses `Account.routerLimitNotifiedMonth` (YYYY-MM string), claimed atomically inside the phase-1 SERIALIZABLE txn so concurrent reloads breaching together send exactly one email. Read-time comparison with the current month re-arms on rollover — no cron. Admin surface: `RouterLimitBanner` (reuses `DelinquencyBanner` shell) above the Router/BYOK tabs in `ModelAccessCard`, with a popover "Adjust limit" form that PATCHes the existing /api/account/[owner]/billing/settings route. Same `assertBillingAdmin` gate that owns the other billing settings — no new auth surface. See wiki/billing.md § Router monthly spend limit for the full contract + edge cases. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(billing): anneal pass on monthly Router spend limit (#660) Round-1 review across 5 lenses (billing-subsystem, correctness, security, operational-readiness, research-validated-assumptions) surfaced one critical + three actionable major findings on top of [#748](https://github.com/pullfrog/app/pull/748). **Critical — CAS never matched NULL.** `claimRouterLimitNotificationSlot` used Prisma `NOT { routerLimitNotifiedMonth: monthKey }`, which compiles to `field != value` — UNKNOWN (not TRUE) against the post-migration `NULL` default. First breach for any account would never claim the slot, never stamp the row, and never fire the email (hard_cap or alert_only). Replaced with `OR: [{ field: null }, { field: { not: monthKey } }]`, mirroring the `maybeNotifyLowBalance` pattern. **Major — email gap on manual-top-up over cap.** Breach email was only wired through `/api/proxy-token`. A manual `/billing-top-up/<owner>` that crosses the cap blocks dispatch via `reserveRun` but never hits proxy-token, so the user got the PR comment but no email. Wired the CAS + `after(maybeNotifyRouterLimit)` into `reserveRun`'s PaywallError catch (the SERIALIZABLE txn rolled back when we threw, so we re-claim with the global client; single-statement CAS is its own race boundary against concurrent proxy-token claims). **Major — PR paywall comment leaked $ figures.** `router_limit` body embedded `($X of $Y)` in a comment visible to anyone with PR read access (public repos, forks, outside collaborators). Other paywall types deliberately avoid amounts. Removed; deep link still points to the authenticated console for the figures. **Medium — observability.** Added `[router-limit]` structured logs at the three enforcement sites (proxy-token hard_cap 402, proxy-token alert_only breach, reserveRun paywall) so on-call can grep "did the cap fire for customer X this month." **Medium — customer docs.** Added a `### Monthly spend limit` section to `docs/billing.mdx` (Mintlify) describing the two modes and the manual-top-up caveat. **Doc — refund/dispute interaction.** Documented in `wiki/billing.md` that the cap inherits the existing webhook semantics: disputed `router_topup` drops from the sum (cap briefly un-trips); refunds don't flip status today so refunded top-ups keep counting. Matches wallet behavior — not redefined here. Accepted as-is (documented or pre-existing): `after()` reliability vs stamp-before-send tradeoff, alert_only email fires before Stripe phase-2, proxy-token reads limit fields outside SERIALIZABLE scope (brief TOCTOU on admin lowering cap), stale paywall comment on cap clear, no global kill switch (per-account `alert_only` flip is the practical kill switch), no audit log on cap changes (no existing audit infra), action version not bumped (separate release commit). Co-authored-by: Cursor <cursoragent@cursor.com> * fix(billing): anneal round 2 on monthly Router spend limit Round-2 anneal (billing-subsystem, correctness, research-validated, user-journey, operational-readiness) surfaced a critical merge conflict and a handful of major correctness + UX gaps on top of [#748](https://github.com/pullfrog/app/pull/748). **Critical — merge conflict.** While #748 was open, [#755](https://github.com/pullfrog/app/pull/755) extracted `formatBillingErrorSummary` from `action/main.ts` to `action/utils/billingErrors.ts`. The PR's new `router_monthly_limit` arm still lived in `action/main.ts`. Took main's slim orchestrator wholesale; moved the arm into the extracted file. **Major — cap = payments only, not dispatch.** `reserveRun` was pre-empting all PR-comment / `/trigger` dispatch on `spent >= limit` regardless of wallet balance, contradicting the cap's positioning as "ceiling on what you pay." An account with $500 of paid-up wallet and a breached $100 cap couldn't trigger any new run via the comment path, while GitHub UI re-runs (which bypass `reserveRun`) succeeded — surface inconsistency. Deleted the pre-dispatch gate; `/api/proxy-token` is now the sole enforcement point, refusing only the next auto-reload that would push past. Wallet credit always drains. Dropped the now-dead `router_limit` arm in `buildPaywallCommentBody`, the dead `routerSpentCents`/`routerLimitCents` fields on `PaywallError.detail`, and the post-paywall email-fire David added — all unreachable. **Major — split `manual_topup` from `router_topup`.** Manual on-session top-ups at `/billing-top-up/<owner>` were landing as `Invoice.kind = "router_topup"` and counting toward the cap. The cap exists to brake *passive* runaway (auto-reload loops); a manual top-up is a deliberate click-through that the user owns. Added `InvoiceKind.manual_topup`, flipped the manual write site + `createTopUpCheckoutSession` metadata, broadened wallet / reconcile / billing-report reads to `kind IN (router_topup, manual_topup)`, and scoped `getRouterSpentThisMonthCents` (the cap aggregate) to `router_topup` only. Worked example: cap=$300, reload=$100 → exactly three reloads succeed; a fourth is blocked. Historical rows stay labelled `router_topup` (no backfill); the asymmetry is small and accepted since the manual flow only existed alongside auto-reload for a brief window. Extended the `invoices_kind_matches_stripe_columns` CHECK so `manual_topup` follows the same shape as `router_topup` (PaymentIntent-backed, no stripeInvoiceId); split into a second migration because PG forbids using a freshly-added enum value in the same transaction. **Major — email reframed around the triggering reload event.** The `alert_only` body was reporting a pre-eager-write `spentCents` while the dashboard reads the post-commit value, so email and dashboard disagreed by exactly one reload. Both flavors now say "Your most recent $50 auto-reload brought you over your $300 monthly limit" instead of a running spent-of-cap total — no reconciliation needed, no more "you've hit your monthly cap" copy firing for partial breaches (spent=$80 of $100, reload=$30 would have triggered that wording). **Major — `/trigger/<owner>/<repo>/<n>` paywall copy.** Hardcoded "You've used your 30 free runs this month. Add a card to continue at 7¢/run." regardless of `detail.reason`. Branched on `cap` vs `delinquent` so each paywall surfaces actionable copy with the right CTA. `router_limit` no longer flows through here (per F4 above). **Major — RouterLimitBanner.** Added an `isAlertBreached` visual state (amber palette) so an `alert_only` account at $240 of $200 no longer renders in the same neutral zinc chrome as a healthy under-cap account. Updated popover copy to reflect the auto-reload-only scope. **Medium — paywall log line.** Added `detail.reason` to the `[Installation X] paywall:` log so on-call grepping for "why was this paused" can distinguish `cap` from `delinquent`. **Cleanup.** Dropped dead `utcMonthKey` import + re-export in `maybeNotifyRouterLimit.ts`. Renamed file-internal `reconcileRouterTopup*` fns + their reconcile-kind labels to `reconcileTopup*` / `topup_*` since they now handle both kinds. Updated wiki/billing.md + docs/billing.mdx + schema doc comments throughout. Co-authored-by: Cursor <cursoragent@cursor.com> * refactor(billing): drop routerLimitNotifiedMonth sentinel; rely on Resend idempotency-key The sentinel was the same anti-pattern as `routerLowBalanceEmailedAt` sitting next to it — a single-purpose state column on `Account` that encoded a date as a string and required a custom CAS predicate to read/write race-safely. Plus it had real holes: Resend send failure left the sentinel stamped and the account silently un-emailed for the month (F11), mode flips mid-month didn't re-arm (F9), and cap-lowered edge cases never fired at all. Replace it with: fire `maybeNotifyRouterLimit` on every breaching reload, let the Resend `Idempotency-Key` `router_limit:<accountId>:<monthKey>:<mode>` collapse repeats inside Resend's 24h dedup window. Continuously-breaching accounts get ~1 reminder per day; brief Resend outages self-heal because the next breaching reload re-attempts the send. Mode is in the dedup key so `alert_only → hard_cap` mid-month re-arms a fresh email with the appropriate copy. Drops `Account.routerLimitNotifiedMonth` and `claimRouterLimitNotificationSlot`; simplifies the proxy-token phase-1 branch significantly. Net diff is negative LOC and the data model loses a single-purpose sentinel. Migration was branch-local — never deployed — so I edited the original add-cap migration in place to drop the column from the ALTER TABLE rather than chain a drop-column migration on top. Preview Neon branches reset automatically on history rewrite per wiki/migrations.md. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(billing): hide RouterLimitBanner when no cap is configured The banner was unconditionally rendered for every billing-enabled account, including pure-BYOK admins who never touch Router. They got "No monthly spend limit / Router has spent $0.00" + a divider as visual noise on the model access page — basically nagging them to set a feature they may not want. Running without a cap is valid; we don't nag. ModelAccessCard now gates the banner block (banner + dividers) on `routerMonthlyLimitCents !== null`. RouterLimitBanner drops the no-limit visual state, the "Set monthly limit" CTA text, and the dead `hasLimit` branching. Cleaner three-state shape (under cap / amber breached / brick breached). Discoverability: no-cap users no longer see a UI affordance to set one. That's deliberate — the cap is a power-user feature documented in docs/billing.mdx. If discoverability becomes an ask, we can add a small inline link inside RouterWalletSection without bringing back the always-visible banner. Resolves the only outstanding finding from cursor bugbot's review of ff5328c (banner-visible-for-byok thread). Co-authored-by: Cursor <cursoragent@cursor.com> * docs(billing): docs/wiki match new "no banner without a cap" reality Pullfrog bot review of f7672ca pointed out the customer docs still told users to "Set the cap from the **Monthly spend limit** banner in the **Model costs** card" — but after hiding the banner for no-cap accounts there is no such banner to use until you already have a cap. Catch-22 for first-time setup. Rewrote docs/billing.mdx to be self-contained: explain what the cap is, what the two modes do, what the banner shows *once configured*, and direct admins to PATCH the billing settings endpoint (or reach out to support) for first-time setup. Cap is positioned as optional; running without one is the documented default. Wiki paragraph in wiki/billing.md updated to match — banner is only rendered when a cap exists, three visual states (under / amber / red), no first-time-setup UI nag by design. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(billing): move monthly cap into RouterWalletSection as a normal settings row; drop the banner entirely The standalone `RouterLimitBanner` was the wrong shape. It only rendered when a cap was already configured (so there was no UI to discover the feature in the first place — first-time setup required hitting the API directly), and it occupied prominent real estate above the tabs to surface state that already lives in the row's own input when the form moves down where it belongs. New shape: monthly cap is just a third row inside `RouterWalletSection` sibling to **Auto-reload amount** and **Auto-reload threshold**. Gated the same way (card on file + auto-reload enabled — the only state where the cap actually means anything). Empty input → no cap, with placeholder "No limit". Setting a number reveals a **Behavior at limit** toggle built on the same `Tabs` slider component used for the Router/BYOK tab switch, so the look matches the rest of the card. Deletes: - `RouterLimitBanner` component (212 lines) - banner mount + conditional + spacers in `ModelAccessCard` - `AlertTriangle` is still imported (used by `DelinquencyBanner`) Adds: - one settings row in `RouterWalletSection` with the cap input + mode tabs - `routerMonthlyLimitUsd` / `routerLimitMode` plumbed through the existing `saveSettings` helper (widened to accept `string | null`) - `Tabs` / `TabsList` / `TabsTrigger` import Docs + wiki updated to match the new shape; the customer doc no longer points at a banner that won't appear. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(billing): split monthly cap input and Behavior-at-limit toggle into separate rows with hr between Previously bundled both into one row block. Restructure: cap input is its own row; Behavior-at-limit Tabs gets a sibling row with the standard `h-5 + hr + h-5` separator between (matching the rhythm of auto-reload amount → threshold → monthly cap). Mode-toggle row is gated on `routerMonthlyLimitCents !== null` so the hr + tabs only appear once a number is in the cap input. Co-authored-by: Cursor <cursoragent@cursor.com> * fix(billing): right-justify Behavior-at-limit Tabs to mirror Auto-reload toggle row Same `flex items-center justify-between gap-3` layout as the Auto-reload row: label group on the left, control on the right. Drops the vertical stack in favour of the horizontal one — looks identical to the toggle row directly above. Co-authored-by: Cursor <cursoragent@cursor.com> --------- Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Colin McDonnell <colinmcd94@gmail.com> * drop italic TL;DR commit-range line from review body the metadata (sha range, commit list, timestamps) is already in the html comment for downstream agents. the visible italic line was clutter and the ellipsis form broke the second sha's auto-link on github anyway. * add agent-browser fallback rule for unreachable chrome devtools mcp * onboarding: gated org-console wizard (#762) * onboarding: gated org-console wizard Replaces the org console's `/console/[owner]` page with a single-card, "growing" stepper when the account has zero `Repo` rows. Walks first-time users through billing mode, BYOK provider+key (if applicable), repo pick, workflow file creation, and a celebratory redeem-credit moment before landing them back on the now-populated org console. ## What's new - New: `components/OnboardingStepper.tsx` — the wizard. Six steps, each derived from real persisted state (Account.modelAccessMode, AccountSecret, Repo). Step state ladder with progressive disclosure and click-to-edit collapsed summaries. - New: `app/console/[owner]/OnboardingView.tsx` — page-chrome wrapper that hosts the stepper inside the same header/sidebar shell as the member view. - Modified: `app/console/[owner]/page.tsx` — adds a `prisma.repo.count` gate alongside existing parallel queries; renders OnboardingView when count === 0, else falls through to the existing repo grid. ## Schema - Flipped `Account.modelAccessMode` default from `byok` to `router`. Router is the lower-friction default (signup credit funds first ~150 runs without a card; users can flip to BYOK explicitly via the wizard or the existing `<ModelAccessCard>` switch). Existing rows keep their current explicit value — Postgres column-default change doesn't backfill, by design. - Migration: `20260516014601_modelaccessmode_default_router`. ## Credit-claim semantics Killed the historical mount-time auto-claim on `<SignupCreditModal>`. All claims are now explicit clicks, fired from one of two surfaces: 1. Wizard step 6 "Redeem $10 credit" CTA (Router branch, eligible). 2. New explicit "Redeem $10 credit" button on `<BillingCard>`'s Router wallet section, visible only when the new server-derived `signupCreditEligible` flag is true (promo active + no prior signup or welcome grant). Covers existing users who'd otherwise lose the auto-claim entry point. `<SignupCreditModal>` is now a controlled component (`open` / `onOpenChange` / `amountCents` props) with a sibling `useClaimSignupCredit(owner)` hook for explicit invocation. The Sparkles celebration dialog rendering is unchanged. ## Other touched surfaces - `app/api/create-workflow/route.ts`: optional `model` body field. When present, the route updates `Repo.model` on the row that `createWorkflowForRepo` just created/surfaced — wizard threads the picked provider's `preferred` model alias through here so a fresh repo doesn't sit on null/auto. - `app/api/account/[owner]/billing/route.ts`: surfaces `signupCreditEligible: boolean` (derived from `SIGNUP_CREDIT_PROMO_ACTIVE` + grant scan). Drives the new explicit redeem button. - `components/AgentSettings.tsx`: fixes the Router-no-billing copy lie ("Runs will draw from your signup credit until exhausted" was false — `isInfraCovered` gates Router minting on `hasCardOnFile`, not balance, so credit-only-no-card users can't actually spend the grant on Router runs). New copy: "Add a card to use Pullfrog Router. Your $10 signup credit (if claimed) applies on top." ## Resume-tomorrow detection Every step's expansion is derived from persisted state (no new column, no localStorage). With the Router default flip, `modelAccessMode === "byok"` is now a reliable signal of explicit user pick, eliminating the heuristic that the byok-default schema would have required. The only ambiguous case is "Router-bailed-before-redeem" (looks identical to a default-Router fresh visit since neither card nor grant exists yet) — acceptable 1-click cost on revisit. ## Testing - `pnpm lint`: clean - `pnpm format`: clean - `pnpm typecheck`: clean - `pnpm -C action test`: 596/596 passing - Visual verification: blocked — Chrome DevTools MCP returned "Not connected" across both available servers. Manual walkthrough needed before merge to confirm step transitions, going-back UX, and the celebration modal redirect destinations match the plan in `.cursor/plans/org_onboarding_stepper_4fdfebbb.plan.md`. * onboarding: drop accordion, multi-repo bulk-onboard, full-width radio rows Three rounds of UX feedback rolled in: 1. **Drop the accordion.** Steps no longer collapse to a one-line summary when "done" — the wizard literally grows by appending steps below as the user progresses, and earlier steps stay fully interactive (re-flip Router→BYOK, re-pick provider, toggle a repo) without any "edit" affordance. `StepShell` now always renders its body for any step the user has reached; the only state distinction is the number circle (filled = active, check = done). 2. **Step 1 is full-width radio rows, not narrow tabs with side-by-side info tiles.** Two rows, each with the option title, an inline "Recommended" badge on Router, and a description sentence inside the row. The persisted `Account.modelAccessMode` (default `router`) drives the initial selection, so step 1 always has one row picked on first paint — no "neither selected" empty state. 3. **Multi-repo bulk-onboard.** Step 4 now uses checkboxes; copy reads "Select the repos you'd like to install Pullfrog into. We'll create a pullfrog.yml GitHub Actions workflow file in each." Step 5 fans out N parallel `POST /api/create-workflow` calls (concurrency capped at 4) and renders per-repo status inline (running → committed / PR #N / already configured / error). Step 6 celebrates with a multi-result headline ("Pullfrog is set up across N repos") and a sub-line breaking down `committed · PRs awaiting merge · failed` plus a per-repo PR list when any PRs were opened. Single- repo path renders the same control surface but with singular copy. Other bits: - Per-step description sentences below every title. - Repo picker shows totalCount inline with the pagination controls and "N repos selected" summary below the table. - Dropped the `userPickedBillingMode` and `editingStep` state machinery + the `isFreshDefault` heuristic — all simplified out by the no-accordion design (we just trust `billingMode` directly). - `createWorkflowPR` PR body already links back to `pullfrog.com/console/<owner>/<repo>` with a "Verify workflow" CTA; no change needed there. * fix(onboarding): provider tile labels — getProviderDisplayName expects slug `getProviderDisplayName` from `pullfrog/internal` parses its argument as a `provider/model` slug. Step 2 was passing bare provider keys (e.g. "anthropic"), which made the helper throw "invalid model slug 'anthropic' — expected 'provider/model'" and crashed the BYOK branch with the page-level error boundary. Replace with a local `providerDisplayName` that reads the registry directly (`providers[key].displayName`). Drops the unused `getProviderDisplayName` import. Caught by Chrome DevTools end-to-end: clicking Bring-your-own-key on the fresh wizard renders the page-level error. Re-verified post-fix: BYOK flow shows step 2 with all 9 provider tiles correctly labeled (Anthropic / OpenAI / Google / xAI / DeepSeek / Moonshot AI / Amazon Bedrock / OpenRouter / OpenCode), step 3 reveals on tile click. Also adds a guardrail to AGENTS.md: don't silently abandon visual verification when DevTools breaks. Recovery is always possible (pkill -9 chrome-devtools-mcp + pkill puppeteer + rm Singleton locks + retry several times); if it genuinely won't recover, abort and tell the user — never mask as "verified by code review". * agents.md: never give up on Chrome DevTools MCP failures Recovery is always possible (pkill chrome-devtools-mcp, remove Singleton locks, retry several times). If genuinely unrecoverable, abort and tell the user explicitly — never silently mask as "verified by code review". Visual verification is non-negotiable for UI changes. * onboarding: polish — checkbox color, redundant labels, copy Caught during chrome-devtools verification of the BYOK + cross-page selection flows: - **Checkbox color**: native browser pink/red replaced with evergreen via `accent-evergreen-600`. Visually consistent with the rest of the wizard's selection states. - **Bedrock provider tile**: was rendering "Amazon Bedrock" twice (provider name + recommended-model name both resolve to "Amazon Bedrock" because Bedrock has no `preferred` model under `providers.bedrock.models` — its single routing entry IS the recommended pick). Suppress the recommended subtitle when it duplicates the provider name. - **Step 6 description**: tightened from a clunky two-clause sentence about workflow file landing to a single direct call: "Mention @pullfrog in any PR or issue to dispatch a run. (Branch-protected repos: merge the PR first.)" - **Wizard intro**: was "Set up Pullfrog for your first repo" — outdated since multi-repo. Now: "Connect Pullfrog to your repos. Each step unlocks the next as you go." Cross-page multi-select also verified: selections from page 1 persist when navigating to page 2 and back. "N repos selected" counter reflects total across all pages. BYOK secret-add flow verified end-to-end: AddSecretModal opens with the env var pre-filled, save triggers secrets refetch, step 3 flips to "✓ ANTHROPIC_API_KEY configured", step 4 reveals automatically. * onboarding: serial install, inline secrets, explicit credit redeem - step 3: replace modal-based secret entry with inline password fields per provider, with deep links to provider dashboards. claude code OAuth surfaces as a distinct group when anthropic is picked. bedrock gets three-field form. github actions secrets path is collapsible with org/personal-aware urls + self-certify. - step 4: merge repo-pick + workflow-create into one step. install is now serial (visible slow-reveal) instead of concurrent. continue button renders immediately on submit, disabled until every repo reaches a terminal state. errored rows render a single soft amber 'failed' label. pagination uses chevron buttons + keepPreviousData (no layout shift). - step 6: explicit 'redeem $10 credit' for router+eligible, 'complete setup' otherwise. final redirect is a hard refresh so the repo grid picks up. - signup credit: drop the mount-time auto-claim modal in favor of explicit user clicks. new useClaimSignupCredit hook + RedeemSignupCreditCallout banner inside RouterWalletSection so a BYOK→Router flip surfaces a one-click redeem affordance. - billing mode is now optimistic (local state + background PATCH) and initialBillingMode + signupCreditEligible eager-load via server props to kill the multi-second click latency. - skip onboarding: header button sets pullfrog_skip_onboarding cookie; server reads it in page.tsx and falls through to the regular grid. - demo mode: NEXT_PUBLIC_ONBOARDING_DEMO=1 cycles the install progress list through pending/running/committed/PR/existing/failed states. - createWorkflowForRepo: PULLFROG_FORCE_PR_CREATION=1 skips direct commit to exercise the PR fallback locally. * onboarding: review feedback — focused eligibility query, best-effort model pre-fill, claim error toast - billing/route.ts + console/[owner]/page.tsx: replace top-N recentGrants scan for signup-credit eligibility with a focused findFirst({ reason: { in: [SIGNUP, WELCOME] } }). the prior query could return any 5/10 rows (no orderBy on page.tsx) and miss a prior signup/welcome grant if a future grant reason (refund/referral/etc.) ever ships. recentGrants stays for the billing-history list. - create-workflow/route.ts: gate Repo.model updateMany on result.type === "created" so an existing user-set model isn't clobbered when the workflow file already exists. wrap in try/catch: GitHub side effect already succeeded, so a transient DB blip shouldn't 500 the route and have the UI report failure on a partially-completed setup. - SignupCreditModal: add onError toast to useClaimSignupCredit so transient redeem failures surface ("Couldn't redeem your credit. Try again in a moment."). callers .catch(() => null) the rejection so it doesn't propagate as an unhandled rejection in the React handler. - OnboardingStepper: trim stale "per-row try again button" wording from progressRef + processRepo comments — that button was removed in the prior commit per design feedback. * router: fix unspendable signup credit on no-card private repos (#791) The bug ------- `run-context/route.ts` gated `proxyModel` minting on `isInfraCovered`, which is `oss || hasCard`. So a no-card account with positive wallet balance (signup credit, top-up, etc.) on a private repo would never get a `proxyModel` set on the run context. The action runtime then fell through to whatever provider keys happened to be in the workflow env — using the user's BYOK keys without their knowledge if any were configured, or failing the run entirely otherwise. Meanwhile `proxy-token/route.ts` already gated correctly on `oss || hasCard || balance > 0`. The two routes disagreed, with run-context being strictly more restrictive, so the agent never even attempted to call proxy-token for these accounts. The wiki at `billing.md:1052` documented the *intended* behavior ("a Router usage row can debit a wallet with no card on file"), aspirational against the actual code. The action side had a parallel bug at `action/utils/proxy.ts:151` — it re-derived `isInfraCovered({ isOss, plan })` and short-circuited mint even when the server set `proxyModel`. Belt-and-suspenders that was strictly more restrictive than the server. Production impact ----------------- Queried 55 router-mode no-card accounts holding signup credit: - ALL have wallet balance = exactly $10.00 (untouched) - ALL have 0 router proxy keys ever minted, 0 hwm usage - ~25 have successful runs (using BYOK env vars from their workflow, unaware their credit isn't being touched) - The rest have zero successes; some accumulated 25+ failures (e.g. `onechannelpe`: 25 failures, 0 successes, no card, $10 credit). The fix ------- - `run-context/route.ts`: widen `useRouter` to match proxy-token's gate. OSS short-circuits as before. Otherwise: router mode + card on file → mint; router mode + no card + positive balance → fetch balance, mint if > 0. Skip the balance read when a card is on file (auto-reload covers it without needing pre-flight balance — keeps the hot path single-query). - `action/utils/proxy.ts`: drop the redundant `isInfraCovered` check. `ctx.proxyModel` IS the signal — the server is the authority on funding decisions; the action just trusts and mints. - `wiki/pricing.md`: correct the Router proxy key minting gate row + add a paragraph explaining why this gate diverges from `isInfraCovered`. - `wiki/billing.md`: rewrite the misleading "proxy-token returns 402" paragraph to describe what actually happens at both routes. `isInfraCovered` is unchanged. It still gates Pullfrog-paid features (learnings writes, indexing). The bug was in conflating "Pullfrog pays for marginal infra" with "user can fund a Router run via wallet" — different concerns, now untangled. * revert: extract router-gate fix into its own PR The router fix at a14bcdd4 is being shipped as a standalone PR so it can be reviewed and merged independently of the onboarding-wizard work. Reverting here keeps #762 focused on the wizard. The fix itself landed at https://github.com/pullfrog/app/pull/792. * router: fix unspendable signup credit on no-card private repos (#792) * router: fix unspendable signup credit on no-card private repos (#791) The bug ------- `run-context/route.ts` gated `proxyModel` minting on `isInfraCovered`, which is `oss || hasCard`. So a no-card account with positive wallet balance (signup credit, top-up, etc.) on a private repo would never get a `proxyModel` set on the run context. The action runtime then fell through to whatever provider keys happened to be in the workflow env — using the user's BYOK keys without their knowledge if any were configured, or failing the run entirely otherwise. Meanwhile `proxy-token/route.ts` already gated correctly on `oss || hasCard || balance > 0`. The two routes disagreed, with run-context being strictly more restrictive, so the agent never even attempted to call proxy-token for these accounts. The wiki at `billing.md:1052` documented the *intended* behavior ("a Router usage row can debit a wallet with no card on file"), aspirational against the actual code. The action side had a parallel bug at `action/utils/proxy.ts:151` — it re-derived `isInfraCovered({ isOss, plan })` and short-circuited mint even when the server set `proxyModel`. Belt-and-suspenders that was strictly more restrictive than the server. Production impact ----------------- Queried 55 router-mode no-card accounts holding signup credit: - ALL have wallet balance = exactly $10.00 (untouched) - ALL have 0 router proxy keys ever minted, 0 hwm usage - ~25 have successful runs (using BYOK env vars from their workflow, unaware their credit isn't being touched) - The rest have zero successes; some accumulated 25+ failures (e.g. `onechannelpe`: 25 failures, 0 successes, no card, $10 credit). The fix ------- - `run-context/route.ts`: widen `useRouter` to match proxy-token's gate. OSS short-circuits as before. Otherwise: router mode + card on file → mint; router mode + no card + positive balance → fetch balance, mint if > 0. Skip the balance read when a card is on file (auto-reload covers it without needing pre-flight balance — keeps the hot path single-query). - `action/utils/proxy.ts`: drop the redundant `isInfraCovered` check. `ctx.proxyModel` IS the signal — the server is the authority on funding decisions; the action just trusts and mints. - `wiki/pricing.md`: correct the Router proxy key minting gate row + add a paragraph explaining why this gate diverges from `isInfraCovered`. - `wiki/billing.md`: rewrite the misleading "proxy-token returns 402" paragraph to describe what actually happens at both routes. `isInfraCovered` is unchanged. It still gates Pullfrog-paid features (learnings writes, indexing). The bug was in conflating "Pullfrog pays for marginal infra" with "user can fund a Router run via wallet" — different concerns, now untangled. * action: drop dead isInfraCovered + plan param post-fix Cleanup the action-side dead code introduced by the previous commit's removal of the redundant `isInfraCovered` re-derivation in proxy.ts: - delete `isInfraCovered` from action/utils/runContext.ts (was the only callsite; mirror in server's utils/billing.ts is unchanged and still load-bearing for learnings/indexing) - drop unused `plan: AccountPlan` param from `resolveProxyModel` / `runProxyResolution` (and the corresponding `AccountPlan` import + the `plan: runContext.plan` arg at the main.ts call site) - update the action/mcp/server.ts comment that pointed at the now-gone action mirror to reference the server-side `utils/billing.ts` instead `AccountPlan` itself is still load-bearing (mcp/server, runContextData, run-context fetch), only `isInfraCovered` and the dead `plan` parameter go away. * eager signup credit + free-OpenCode fallback when BYOK has no key (#789) * eager signup credit + free-OpenCode fallback when BYOK has no key addresses the silent-churn pattern that took out 15 first-run-failure accounts post-launch: GH Actions secret references resolved to empty strings (because the secrets didn't exist on the repo), the action launched Claude Code with no key, the LLM provider 401'd, and the run died in seconds with a synthetic "Invalid API key" message. those accounts had no Router credits to fall back to because the lazy claim required a dashboard visit they never made. three changes, one PR: 1. Eager $10 signup credit at account creation. Both account-creation sites (`upsertAccountByClerkId` for dashboard signin, `fetchOrCreateRepo` for CLI / GH-App-only) now insert the `CreditGrant { reason: "signup" }` in the same transaction as the `accounts` row. CLI installers who never sign in get the credit. The dashboard `/signup-credit/claim` POST stays as an idempotent backstop for accounts created before this shipped. 2. Free-OpenCode fallback in the action. When the configured BYOK slug needs a provider key the runner doesn't have, swap to `opencode/minimax-m2.5-free` before agent selection so the run still succeeds. Surfaced via a `» fell back from <slug> to <free>` warning in the action log. Skipped on Router runs (Pullfrog mints the key) and when no model is configured (auto-select-with-throw still fires for the genuinely-misconfigured case). 3. New action-test fixture `byok-no-keys-fallback` that empty-strings every known provider key (matching how GH Actions handles missing secrets) and asserts the run succeeds with the fallback log line present. plus a unit test for the helper covering each skip case. skipping the schema flip from `byok` to `router` — that's coming via the onboarding-stepper PR (#762). * fallback: skip Bedrock + surface in PR-comment footer addresses copilot review on #789 (real bug — parseModel throws on Bedrock raw IDs that have no slash, would crash before validateBedrockSetup could surface its own error) and the user-side ask to make the fallback visible in PR comments. - selectFallbackModelIfNeeded skips when resolvedModel has no '/' so Bedrock routing IDs (e.g. us.anthropic.claude-opus-4-7) don't crash inside hasProviderKey -> parseModel. unit test covers it. - toolState.modelFallback records the configured slug we fell back from. set in main.ts when fallback engages. - buildPullfrogFooter accepts fallbackFrom and renders "Using `MiniMax M2.5` (free) (credentials for Claude Opus not configured)" so the substitution is visible in PR comments, reviews, PR bodies, and error reports. - threaded through all four action-side footer call sites (mcp/comment, mcp/pr, mcp/review, utils/errorReport). server-side call sites in triggerWorkflow.ts / handleWorkflowRunWebhook.ts fire pre-action and don't have toolState — left as-is. * fallback footer: use provider display name + document email asymmetry addresses pullfrog reviewer findings on #789: - footer now shows 'credentials for Anthropic not configured' (provider display name from `providers.anthropic.displayName`) instead of the per-model name. credentials are provider-scoped (ANTHROPIC_API_KEY covers all Anthropic models), so this matches what the user actually needs to fix. - document the intentional asymmetry between eager and lazy signup credit paths: eager skips both the signupCreditClaimedEmail and the per-grant team@ alert. comment explains why (the 'new account created' alert already covers it on the eager path; the user-facing email assumes a user-initiated action that hasn't happened yet for CLI/GH-App-only signups). - skipping the backfill for the 15 historical accounts per user's earlier decision — they all uninstalled, so the cohort self-selected out of being reachable. * fallback: gate on resolvedModel + skip resolveModel re-resolve post-swap local agnostic fixture run surfaced two real bugs the unit tests didn't catch: 1. fallback gate was on configuredSlug (=payload.model) but the test uses PULLFROG_MODEL to set the model, which is read by resolveModel AFTER its slug arg. configuredSlug stayed undefined → fallback never fired. drop configuredSlug from the helper signature; gate purely on resolvedModel since that's the same value regardless of how the model was specified (DB config vs PULLFROG_MODEL env). 2. when fallback engaged, the post-swap resolveModel({slug: fallback.to}) call was ALSO honoring PULLFROG_MODEL, re-overriding the fallback target back to the unkeyed model. validateAgentApiKey then threw "no API key found" against the original model. fix: skip the re-resolve. fallback.to is already a CLI-ready specifier. unit tests updated for the new helper signature (8 tests, all pass). fallback log line confirmed emitted in the local run pre-second-fix; the second fix unblocks the validation that previously threw. * models-bump: harden CI and bot prompt against catalog hallucinations PR #790 (the first bot-authored models-bump PR) shipped a broken bump for openrouter/gemini-flash: the bot pattern-matched the parallel google bump and fabricated `openrouter/google/gemini-3.5-flash`, which exists on the OpenRouter API but not on models.dev's openrouter section — the catalog OpenCode actually reads. The slug failed at runtime with `Model not found`. Two CI gaps let it through: 1. `models-live` matrix pruned every `openrouter/*` and keyed `opencode/*` alias as a "passthrough", smoke-testing only one canary per routing layer. But those aren't passthroughs — each is a distinct models.dev catalog entry that can drift independently of the direct-provider mirror. Drop the pruning; smoke every keyed alias (53 jobs, up from 25). Only `bedrock/byok` stays pruned (sentinel resolve). 2. `models-catalog` test (the integrity gate that asserts every resolve exists on models.dev) was main-only by design — to keep upstream catalog churn from blocking unrelated PRs. But it's exactly the test we want running on the bot's own catalog edits. Add `pullfrog/models-bump` head-ref to its trigger. Also tighten the bot prompt in models-bump.yml: new rule 0 requires every new `resolve` to equal `<alias-provider>/<c.modelId>` for some `c` in the alias's own `candidates[]` in models-bump-context.json — the deterministic preprocessor only emits candidates sourced from models.dev's mirror, so this gates against the cross-alias pattern-matching that broke PR #790. For `openRouterResolve` the gate is `openRouterCandidates[]` (OpenRouter API), which is necessary but not sufficient; the `models-catalog` job is the authoritative models.dev check. Verified locally: - baseline `pnpm -C action test:catalog` passes 133 tests - simulated the PR #790 hunk (sed'd `openrouter/google/gemini-3.5-flash` into action/models.ts) and the catalog test fails with the right assertion: `model "google/gemini-3.5-flash" not found under openrouter on models.dev` - `FULL=1 node action/test/matrix.ts` emits 53 aliases (was 25); every openrouter/* alias and every keyed opencode/* alias now smoked --------- Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Colin McDonnell <colinmcd94@gmail.com>	2026-05-20 03:58:18 +00:00
pullfrog[bot]	0abaaa1e37	chore(oss): add yamcodes/arkenv to OSS program (#776) ... * chore(oss): add yamcodes/arkenv to OSS program * fix(test): strip CODEX_AUTH_JSON in apiKeys auto-select test The beforeEach strip list omitted CODEX_AUTH_JSON, which is in `knownApiKeys` via the openai provider's managedCredentials. When the env has CODEX_AUTH_JSON set, the auto-select "throws when no provider keys are present" assertion finds it and fails to throw. --------- Co-authored-by: pullfrog[bot] <226033991+pullfrog[bot]@users.noreply.github.com>	2026-05-17 20:26:26 +00:00
Colin McDonnell	d495f0b984	surface BYOK failures + chronic-failures card + WorkflowRunStatus mirrors GitHub conclusions (#722) ... - Migrates `WorkflowRunStatus` from `running | completed | cancelled` to a 9-state mirror of `workflow_run.conclusion`. Backfill: old `completed → success`, `cancelled → failure`. New rows write `hook.workflow_run.conclusion` verbatim via `statusFromConclusion`. - Adds Discord links to `formatApiKeyErrorSummary` (both missing-key and 401 invalid-key shapes). - Repo console: `<ChronicFailuresCard>` fires when the last 3 terminal-state runs are all `failure`. Pure DB read; latest-run button hidden for pre-dispatch failures (`runId: null`). - `StatusIcon` distinguishes `cancelled` (gray X, intentional stop) from `failure` (red X) so the visual matches the chronic-card threshold. - Pre-dispatch failures (workflow lookup miss, dispatch API error) write `failure` instead of `cancelled` so they feed the card. - Cascade: every `status: "completed"` filter in billing routes / cron / cohort queries / analyzer becomes `status: "success"`. Verified end-to-end on `pullfrog/preview-722-failure-surfaces` — Better Stack logs confirm webhooks reached the preview deploy and all three e2e runs got `marked as failure (conclusion=failure)` via the new mapper. Closes #679, #702.	2026-05-14 02:39:41 +00:00
Colin McDonnell	8f9208bd3f	feat: Amazon Bedrock support via routing slug (#720) ... * add Amazon Bedrock as a routing slug introduces a single `bedrock/byok` catalog entry that the harness translates to the appropriate Bedrock model ID at run time via `BEDROCK_MODEL_ID`. routes Anthropic IDs through claude-code (with `CLAUDE_CODE_USE_BEDROCK=1`) and everything else through opencode's `amazon-bedrock` provider — keeps the catalog flat for an audience that needs version pinning rather than aliasing. accepts either `AWS_BEARER_TOKEN_BEDROCK` or `AWS_ACCESS_KEY_ID` + `AWS_SECRET_ACCESS_KEY` for auth; both validated alongside `AWS_REGION` and `BEDROCK_MODEL_ID` in `validateAgentApiKey`. catalog drift tests, the bumps cron, and per-alias smoke scripts all skip routing slugs since there's no fixed `resolve` to validate. docs/bedrock.mdx walks through setup; wiki/model-resolution.md has a section explaining why bedrock breaks the usual alias pattern. closes pullfrog/pullfrog#40 * ci: add bedrock env vars to test workflows mirrors the new bedrock provider's required env vars (AWS_BEARER_TOKEN_BEDROCK inherited from org secret + AWS_REGION + BEDROCK_MODEL_ID hardcoded) into both .github/workflows/test.yml files so the ci.test "env vars cover all provider API keys" assertion passes. * docs(bedrock): clearer setup flow + screenshot of model selector restructures the setup section into three concrete steps in execution order: select Bedrock from the dropdown, store the bearer token as a secret (Pullfrog or GitHub — links to keys.mdx for the trade-off), then add region + model id directly in pullfrog.yml since neither is sensitive. enable-model-access in the Bedrock console moved to step 4 (only required once per model and only when AWS rejects the call, not blocking on first run). adds a screenshot of the console model selector with Amazon Bedrock selected so readers can recognize the UI state they're aiming for. * fix(bedrock): tolerate raw Bedrock model IDs in validateAgentApiKey main.ts passes the resolved model into validateAgentApiKey (`payload.proxyModel ?? resolvedModel ?? payload.model`). For Bedrock, `resolveModel` translates `bedrock/byok` into the raw AWS model ID (e.g. `us.anthropic.claude-opus-4-6-v1`), which has no `/` and so trips parseModel inside getModelEnvVars. Detect the no-slash case and re-run the bedrock setup check (auth + region; BEDROCK_MODEL_ID is already enforced upstream by resolveModel). Caught by PR #720 e2e dispatch on pullfrog/preview-720-bedrock — "invalid model slug 'us.anthropic.claude-opus-4-6-v1' — expected 'provider/model'". Two regression tests cover the raw-ID path. * fix(bedrock): always prepend amazon-bedrock/ prefix when bedrock-routed opencode.ts was gating the prefix-injection on `!isBedrockAnthropicId(rawModel)`, on the theory that Anthropic Bedrock IDs always go through claude-code. But `PULLFROG_AGENT=opencode` is a documented escape hatch — when it forces opencode for an Anthropic Bedrock model, the prefix still has to be added or opencode fails with 'Model not found: <modelId>/.'. The Anthropic-vs-other discriminant only belongs in resolveAgent. Once an agent is selected, it should consistently honor the bedrock route. Caught by the PULLFROG_AGENT=opencode + Opus 4.6 e2e on pullfrog/preview-720-bedrock — run 25823437606. * ui+docs(bedrock): bespoke setup callout + clearer docs UI: - BedrockSetupCallout in components/AgentSettings.tsx covers both the Model costs section and the onboarding card. Detects bedrock via resolveDisplayAlias().routing === "bedrock", shows a dedicated message ("store AWS_BEARER_TOKEN_BEDROCK as a secret, then put AWS_REGION + BEDROCK_MODEL_ID directly in pullfrog.yml") + link to the setup guide. Replaces the generic "X, Y, or Z is required" prompt that misrepresented the three values as three separate secrets to add (and used the wrong "or" connector for what's actually an AND). - OnboardingCard re-uses the same callout with the gradient-card variant. Docs: - Drop the obsolete "Enable model access" step. AWS retired the manual enrollment page; foundation models auto-enable on first invocation. Anthropic models still need a one-time use-case form for first-time users — surfaced under the AccessDenied troubleshooting entry. - Drop the "Testing a different model in one run" PULLFROG_MODEL note. It introduced the secrets-vs-vars distinction we want to keep out of the bedrock setup story. - Step 3 already recommends hardcoding region + model id in pullfrog.yml. Workflow template: - The default pullfrog.yml customers receive (utils/github/pullfrog.yml.ts) now references AWS_BEARER_TOKEN_BEDROCK from secrets but inlines AWS_REGION and BEDROCK_MODEL_ID as plain values. Matches the docs. * fix(bedrock): three review-caught edges in routing + UI copy Addresses three real issues from PR #720 review: 1. agent.ts: PULLFROG_MODEL=bedrock/byok no longer leaks the literal sentinel "bedrock" downstream. resolveCliModel returns the alias's resolve field verbatim, which for routing entries IS the sentinel. Refactored both the env-override and slug-lookup paths through a shared resolveSlug() that recognizes routing aliases and defers to their backing env var (BEDROCK_MODEL_ID). 2. models.ts: isBedrockAnthropicId() now anchors on a discrete dot/slash/colon-segment match (case-insensitive) instead of a substring contains. The substring check was fragile in both directions for inference-profile ARNs (BEDROCK_MODEL_ID accepts ARNs per AWS docs) — a non-Anthropic profile whose user-chosen name contained "anthropic" would mis-route to claude-code, and an Anthropic profile whose name omitted it would miss CLAUDE_CODE_USE_BEDROCK=1. 3. AgentSettings.tsx: BedrockSetupCallout's configured-state copy showed "AWS_BEARER_TOKEN_BEDROCK configured" even when the user satisfied the gate via AWS_ACCESS_KEY_ID + AWS_SECRET_ACCESS_KEY, gaslighting access-key users about a secret they never set. Detect which auth method is actually present and name the right secret(s) in the success message. Regression tests in models.test.ts (5 new isBedrockAnthropicId cases including positive and negative ARN forms) and agent.test.ts (2 new PULLFROG_MODEL=bedrock/byok cases). 171/171 action tests pass. * yml template: add commented AWS access-key alternative for Bedrock auth Mirrors the IAM access-key path verified end-to-end on PR #720 e2e run 25830764987. Bearer token stays as the primary nudge; the access-key pair is the fallback for users who can't mint Bedrock API keys. * yml template: drop redundant 'or, alternatively' annotation * ui+docs(bedrock): rewrite callout copy + refresh screenshot Reframes the BedrockSetupCallout away from generic BYOK language to a Bedrock-specific message: leads with "Amazon Bedrock is configured entirely via environment variables", lists all four (auth, region, model id), and ends with the requested CTA sentence ("click below to learn more about Bedrock support in Pullfrog"). Promotes the "Bedrock setup guide" docs link from an inline anchor to a prominent button (always visible, regardless of auth state). The "Add AWS_BEARER_TOKEN_BEDROCK" affordance is now a secondary chip shown only when no auth secret is configured. Refreshes docs/images/model-selector-bedrock.png to capture the new callout — the prior screenshot still showed the old generic "BYOK / X, Y, or Z required" wording.	2026-05-14 02:12:38 +00:00
Colin McDonnell	d5d8a0d7ac	fix(#691): drop opencode/gpt-5-nano + opencode/mimo-v2-pro-free (not actually keyless on Zen) (#695) ... * remove opencode/gpt-5-nano and opencode/mimo-v2-pro-free from catalog #7 delete aliases. both were listed as `isFree: true, envVars: []` but neither is keyless on opencode zen, producing a hard-fail `UnknownError: Model not found: opencode/<id>` on every run without an opencode_api_key. fixes pullfrog/app#691 (5 runs across 3 repos, 100% failure rate in the last 24h). root cause: opencode's provider gate (`packages/opencode/src/provider/provider.ts` `opencode:` loader) keeps a zen model only when models.dev reports `cost.input === 0` for it, then signs requests with `apiKey: "public"`. paid zen models get deleted from the autoloaded set and opencode surfaces the deletion as "model not found". - `opencode/gpt-5-nano`: models.dev reports `cost: {input: 0.05, output: 0.4, cache_read: 0.005}`. paid → requires `OPENCODE_API_KEY`. - `opencode/mimo-v2-pro-free`: free on models.dev but not in `https://opencode.ai/zen/v1/models` — zen never served it, so even the public-key path fails. remaining free aliases (`opencode/big-pickle`, `opencode/minimax-m2.5-free`) both pass both checks (cost.input === 0 in models.dev AND present in zen's served list) and continue to work without a key — verified against the opencode source. callers swept: `action/utils/apiKeys.test.ts`, `action/models.test.ts`, `action/test/list-aliases.ts`, `action/test/model-smoke.ts`, `components/ModelSelector.tsx` (`modelIdToUpstream`), `wiki/model-resolution.md`, `wiki/models-catalog.md`. wrote up the free-zen verification rule in models-catalog so the next maintainer can sanity-check both conditions before adding any `isFree` alias. users with a stored `opencode/gpt-5-nano` or `opencode/mimo-v2-pro-free` will now fall through `resolveCliModel → undefined` into the auto-select path — a strict improvement over today's hard fail. no DB migration needed; the slugs are simply unknown and treated like any other unrecognized stored value. * rework: keep mimo deprecated, demote gpt-5-nano to paid, add free-zen invariants revised approach after the first commit over-corrected. mimo was never broken at runtime — `fallback: "opencode/big-pickle"` already routes stored values through to a real free model before any zen call. the literal `opencode/mimo-v2-pro-free` being absent from zen's served list is irrelevant because `resolveCliModel` walks the chain first. restoring it as-is. the actual bug was `opencode/gpt-5-nano`: marked `isFree: true, envVars: []` but `models.dev` reports `cost: {input: 0.05, output: 0.4}` on the opencode provider, so opencode's keyless gate (`packages/opencode/src/provider/provider.ts` `opencode:`) deletes it when `OPENCODE_API_KEY` is missing and the run hard-fails with `UnknownError: Model not found: opencode/gpt-5-nano`. demoting it to a regular paid zen alias (drop `isFree`/`envVars: []`, add `openRouterResolve: "openrouter/openai/gpt-5-nano"` — verified to exist on openrouter at the same price). users without `OPENCODE_API_KEY` now get our explicit "no API key found" error pointing at the secrets page instead of opencode's cryptic upstream error. confirmed via `https://opencode.ai/zen/v1/models` that zen serves no free GPT variants, so there's no cheaper-than-`gpt-mini` free option to suggest in its place. CI gap analysis (why this slipped through): - `models-catalog.main.test.ts` only checked existence + `status !== "deprecated"` on models.dev. paid-model-marked-free regressions and zen-served-list drift both passed. - `models-live` (`model-smoke.ts`) runs with `OPENCODE_API_KEY` in env, so the keyless deletion gate never fires. `gpt-5-nano` returned "OK" in CI even though end users hit a hard fail. - `model-smoke.ts` walks the fallback chain, so mimo would have been smoked as big-pickle anyway — the dead resolve target was never exercised directly. (this is the right design; the gap is at the catalog layer, not the smoke layer.) new tests: - PR-blocking, static (`action/test/models.test.ts`, `isFree invariants`): every `isFree` alias must live under `opencode`, have `envVars: []`, omit `openRouterResolve`, AND have a fallback chain whose terminal alias is also `isFree` (catches "deprecate a free alias to a paid target" — the worst silent-charge regression). - main-only, network (`action/test/models-catalog.main.test.ts`, `opencode Zen served list`): every alias whose terminal-fallback resolve is `opencode/*` must appear in `https://opencode.ai/zen/v1/models`. catches zen dropping a model from its served list. - main-only, network (same file, `isFree models.dev cost`): every `isFree` alias's terminal-fallback resolve must have `cost.input === 0` in the `opencode` provider block on `models.dev`. would have caught `gpt-5-nano` at the next models-bump run. both network tests dedupe on terminal resolve, so deprecated aliases sharing a target aren't double-counted. `pnpm vitest run`: 113 static tests pass. `pnpm test:catalog`: 142 network tests pass against the live `models.dev`, `openrouter.ai`, and `opencode.ai/zen/v1/models` endpoints. wiki/models-catalog.md: rewrote the new "Free-Zen aliases need Zen-side verification" section to (a) describe the two conditions, (b) note that a fallback to an isFree alias is the legitimate escape hatch (mimo's pattern), and (c) point at the three tests by name so the next maintainer can find the enforcement surface. wiki/model-resolution.md points at the new section. * make gpt-5-nano a deprecated free alias falling back to big-pickle revising the previous "demote to paid" approach. the user-facing ergonomics are cleaner: anyone who picked gpt-5-nano under the "Free" badge gets transparent-upgraded to a real free model (big-pickle) instead of suddenly being asked to set OPENCODE_API_KEY. matches the existing mimo pattern exactly. the dropdown already filters `!a.fallback`, so the slug disappears from the picker on its own and the trigger renders it as "Big Pickle" via `resolveDisplayAlias`. no other catalog or test surface changes — the isFree invariants and the main-only zen/cost checks still pass (gpt-5-nano's terminal is now big-pickle, which is both isFree and zero-cost on models.dev, deduping with big-pickle's own row in both network tests). * revise: keep gpt-5-nano as paid alias, backfill affected DB rows instead dropping the deprecated-alias approach. `opencode/gpt-5-nano` is a legitimate cheap paid model people may want with BYOK (`OPENCODE_API_KEY`) — giving it `fallback: "opencode/big-pickle"` would foreclose that for everyone going forward. correct fix is two parts: (a) reclassify in the catalog as a regular paid OpenCode alias: - drop `isFree: true` and `envVars: []` so the local validator demands `OPENCODE_API_KEY` - add `openRouterResolve: "openrouter/openai/gpt-5-nano"` to satisfy the completeness test and route BYOK-via-OpenRouter users - no `fallback` — slug stays visible in the picker as a paid option (b) one-shot DB backfill of provably-affected repos (`scripts/backfill-gpt5-nano-affected.ts`). scope: - `Repo.model = "opencode/gpt-5-nano"` - AND at least one `WorkflowRun` with `inputTokens IS NULL` (evidence of an attempted run that didn't get past the model-init gate) skipped intentionally: - repos whose runs have `inputTokens > 0` — they have a key, gpt-5- nano works for them - repos with zero WorkflowRun rows — never dispatched; touching them would be presumptuous - `LearningsRevision.model` — audit trail of which model authored a revision, rewriting it would falsify history ran against .env.prod: 2 repos stored the slug; 1 was provably affected (sodown4thecause/seobot, 5/5 zero-token runs — matches #691's 3 failed runs from this repo plus 2 outside the 24h audit window). 1 was an internal test account that never dispatched (left as-is). applied: 1 row updated. confirmed idempotent on re-run. the other two repos in #691 (Nantiee/ALTA-breast-pump-tool, keksiqc/ansible-setup-linux) don't store the slug in `Repo.model`; their failed dispatches passed the model inline in the `workflow_dispatch` `prompt` payload, so the catalog fix alone (no longer offering it as free) is what helps them. tests: - models.test.ts: `getModelEnvVars("opencode/gpt-5-nano")` now returns `["OPENCODE_API_KEY"]`, moved into the keyed-model group - apiKeys.test.ts: added "throws without OPENCODE_API_KEY" case - isFree invariants from the previous commit still pass — gpt-5-nano no longer triggers them since it's no longer isFree - main-only catalog tests still pass (gpt-5-nano served by Zen, just paid; no isFree cost check applies) * docs: drop stale GPT Nano + MiMo V2 Pro from free-tier lists addressing pullfrog auto-review feedback on #695. three mintlify pages still advertised both as keyless after the catalog pivot, which now makes the docs affirmatively wrong rather than merely stale: - gpt nano is paid in the catalog (no `isFree`, inherits `OPENCODE_API_KEY`); a user following the docs would hit the same "missing API key" failure that's described 4 lines below in `docs/keys.mdx`. - mimo v2 pro is hidden from the picker (`fallback` triggers `ModelSelector`'s `!a.fallback` filter); the alias only exists for legacy stored-value resolution. a user reading the docs cannot actually pick it. surviving picker-visible free set: Big Pickle and MiniMax M2.5. - `docs/keys.mdx`: drop both bullets from the "Free models" list - `docs/billing.mdx`: drop both bullets from the "Free models" list - `docs/getting-started.mdx`: collapse the inline mention from a 4-model list to "Big Pickle and MiniMax M2.5" * address third review: picker grouping + backfill classifier honesty i had not pulled the third pullfrog review (`02:17:28Z`) when i declared reviews triaged after the docs sweep — the fourth review flagged that three findings remained pending. addressing them now. 1. picker grouping for now-selectable paid gpt-5-nano. when i removed `"gpt-5-nano": "OpenAI"` from `modelIdToUpstream` in the previous pivot-to-paid commit, i mistook it for dead code. it's not — the map IS consulted for paid opencode aliases via `groupByUpstream → getUpstreamLabel` inside the OpenCode submenu's `renderSubContent`. without the entry, `gpt-5-nano` falls back to `getProviderDisplayName("opencode")` = "OpenCode" and gets dropped into its own sub-header instead of joining opencode/gpt, opencode/gpt-pro, opencode/gpt-mini under the "OpenAI" upstream group. re-added with an explanatory comment so the next refactor doesn't make the same mistake. 2. JSDoc / code mismatch in `scripts/backfill-gpt5-nano-affected.ts`. the JSDoc said "at least one `WorkflowRun` with `inputTokens IS NULL`" but the code is `no WorkflowRun has inputTokens > 0` — a strictly broader filter (catches `null` AND `0`). rewrote the scope block to describe what the code actually does, with the operative classifier spelled out: "a billable run with `inputTokens > 0` is proof the agent successfully reached and called the model". 3. classifier breadth (raised in the same review). honest answer: the "no positive-token run" filter IS a heuristic — a repo whose only dispatches happened to fail or cancel for unrelated reasons would get false-positive-classified A. for THIS one-shot population (2 repos, 1 with 5/5 zero-token runs — strong systematic-failure signal) the heuristic was good enough and the dry-run inspection confirmed before APPLY. for any larger reuse of this pattern, you need to cross-reference the runtime error string (`UnknownError: Model not found: opencode/gpt-5-nano`) from GitHub Actions logs or Better Stack — that error doesn't live on `WorkflowRun` rows. added a "Classifier limitations" section to the JSDoc making this explicit. nothing about the actual applied backfill changes — the prod write (1 repo: sodown4thecause/seobot → opencode/big-pickle) is unchanged and re-running the script remains idempotent.	2026-05-13 02:43:08 +00:00
David Blass	b835d53d83	add /anneal + pullfrog-reviewer named subagent + Build self-review polish (#550) ... * cherry-pick updated /anneal command from billing branch + add as Claude Code slash command mirrors origin/billing:.cursor/commands/anneal.md (commit 4f389a8f) into both .cursor/commands/ and .claude/commands/ so the parallel-lens annealing prompt is available in both editors. content is identical between the two files. * anneal: drop REVIEW.md pointer, surface-agnostic dispatch wording, fix modes.ts self-review contradictions Anneal pass over the /anneal slash command and the Build-mode self-review step: - Drop REVIEW.md references in both anneal.md copies. The file does not exist on the Claude Code surface (only .cursor/commands/), and its contents (correctness/security/impact framing) directly contradict the prescribed single-lens, no-pre-shaping discipline. - Replace "Task tool calls" with surface-agnostic "parallel subagent calls" so the meta-prompt does not couple to either CLI's tool naming. - Hedge the "verify via web search" instruction to acknowledge subagents may not have web search available. - modes.ts: drop "and the changed files" — the same step's don't-list forbids handing subagents a curated reading list (in-file contradiction). - modes.ts: restore the "skim only, don't pre-review" warning that the long-form treats as load-bearing. - modes.ts: drop "NO MCP tools" — overbroad; the actual safety property is captured by "no writes, no shell commands, no side effects". Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal: two-round self-anneal of /anneal + modes.ts self-review Expand the multi-lens parallel-review protocol with fixes surfaced by running /anneal on this branch twice. Material additions: /anneal canonical (.claude/commands/anneal.md + .cursor mirror): - promote orientation-vs-defect-hunting distinction to a load-bearing framing in the opening paragraphs - add an empty-target early exit ("nothing to anneal" stop) at §1 - spell out the read-only constraint with the no-op-if-reverted test, and forbid recursive subagent dispatch (incl. agentic MCP tools) - add cleanup-and-debt sub-categories (env vars, feature flags, dangling symbols), supply-chain, test-integrity lenses to the catalog - §1 lens-count rule: explicit trivial/typical/high-risk tiers; "treat as typical" tiebreaker for the unsure case - §2 example uses bare `git diff <primary-branch>` to capture uncommitted edits (three-dot syntax is committed-only) - §5 targeted-follow-up cross-references the fresh-eyes carve-out in Delegation discipline - final-message format spells out coverage shape, findings-table shape, dry-run fix-plan branch, and plan/doc summary branch - stopping criteria distinguish "trivial" from "small / low-risk" action/modes.ts Build mode step 4 (self-review one-pass anneal): - empty-diff early exit; "step 4 mandatory whenever there is a diff" resolves the prior contradiction with the always-runs assertion - lens count by risk (2-3 typical / 4 high-risk single-round-cap / exactly 1 trivial) with separate Tiebreaker - expand swap-in lens menu (research-validated assumptions, security, user-journey, ops, integration, test integrity, supply chain, performance, holistic) so the catalog is a starting menu, not a closed set - rename `cleanup & scope` to `diff hygiene` to avoid colliding with the canonical's broader `cleanup & debt` - delegation discipline bulletized (don't lens-review yourself, don't summarize, don't curate, don't pre-shape, don't mention other lenses); independence rationale stated inline - explicit research-discipline reminder for any lens that touches external contracts (web search, quote URLs) - comment block enumerates deliberate omissions vs the canonical (dry-run, severity categorization, read-only shell) and the deliberate scope decision (sibling diff-producing modes stay solo) action/modes.ts Review + IncrementalReview subagent-dispatch wording: - propagate the no-recursive-dispatch rule (was missing) Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * add set_plan/get_plan + restructure Review/IncrementalReview as parallel-subagent orchestrators Build mode's self-review and Review/IncrementalReview now follow the multi-lens parallel-subagent fan-out pattern from the canonical /anneal protocol. New set_plan/get_plan MCP tools (orchestrator-only) persist the implementation plan in tool state so the self-review's plan-adherence lens can verify the diff against the original intent rather than reconstructing it post-hoc. Subagent "read-only / no further dispatch" is currently enforced via prompt prose only — neither claude-code's --disallowedTools nor opencode's per-agent tools allowlist is configured to scope subagent MCP access. Documented as a deferred ~30-50 LOC follow-up in the modes.ts header comment. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * revert Review/IncrementalReview mode prompts to main; keep Build self-review changes E2e testing on this branch only exercised the trivial-1-lens path for Review (preview repo had only docs PRs). Multi-lens Review fan-out was never directly validated against a real code PR. Splitting the Review/IncrementalReview restructure to its own branch (review-mode-orchestrator, draft PR #555) pending focused validation. Keep on this branch: - set_plan/get_plan MCP tools - Build mode multi-lens self-review (Test 3 directly validated 2-subagent parallel fan-out on a 2-file diff) - /anneal command updates (.claude/ and .cursor/ mirrors) Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * require plan parameter when selecting Build mode Adds an arktype .narrow on SelectModeParams that rejects select_mode({mode:"Build"}) unless a non-empty 'plan' string is also provided. When valid, the plan is stored into ctx.toolState.plan at mode-selection time, so step 4's plan-adherence lens always has a comparison target. This closes the e2e finding that agents never reached for set_plan on their own (5 of 6 runs in production). Build mode prompt updated to reflect that plan is already populated at mode selection; set_plan remains as the mid-task replan tool. Other modes are unaffected. Validation surfaces the error to the agent with a descriptive message including the path ('plan') and recovery instructions, so a failing call is recoverable on the next turn rather than a hard fail. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * move Build-mode plan-required check from arktype .narrow to execute() arktype .narrow predicates aren't JSON-Schema serializable — FastMCP's toJsonSchema() emitted a {code: "predicate", predicate: Function} object instead of a serialized schema. Effect: agents couldn't see select_mode in their tool list (verified by 5 consecutive runs across two models silently bypassing select_mode entirely after the prior commit). Fix: keep the param schema clean (.narrow removed) and check selectedMode.name === "Build" && !params.plan in the execute() body, returning a structured error response. The agent now sees select_mode normally, gets a clear actionable error if it forgets the plan, and can recover on the next turn by retrying with the plan included. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * flip lens architecture: Build = single fresh-eyes subagent, Review/IncrementalReview = multi-lens Build mode self-review previously fanned out 1-4 lenses on the agent's own diff. The bias-mitigation argument for fan-out is weaker for self-review than for reviewing someone else's PR — the orchestrator just wrote the code, so what matters is one fresh-eyes subagent that doesn't share the implementation context, not breadth across parallel angles. Build now dispatches exactly one subagent that gets the original user request and the diff and evaluates whether the diff fulfills the request. Review and IncrementalReview now use the multi-lens orchestrator pattern (triage → parallel read-only fan-out → aggregate → draft comments → submit). For someone else's PR, parallel lenses (correctness, security, research-validated, user-journey, etc.) provide breadth that a single subagent can't carry coherently. Was previously parked on the review-mode-orchestrator branch (PR #555). Removes set_plan/get_plan MCP tools, ToolState.plan field, and the plan parameter on select_mode. Validated end-to-end that those didn't cause agents to actually use plan tracking (5 of 6 e2e runs skipped them); the original user request from the prompt body is the source of truth and the orchestrator already has it. Drops timeout test plan-param workaround that was added for the prior validation. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * split Review/IncrementalReview multi-lens back out to review-mode-orchestrator branch The multi-lens orchestrator restructure for Review/IncrementalReview was bundled into this branch in commit e964ae0c, but it hasn't been validated against a real code-heavy PR (the e2e exercised it only on docs PRs). Splitting it back out keeps this branch focused on the validated half — Build → single fresh-eyes subagent — and lets the Review changes ship in a focused PR (#555 reopened). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal: fix Build prompt contract bugs found by 3-lens review Major fixes: - checkout_pr returns the field as `base`, not `baseRef` (per checkout.ts:611-616). The prompt was telling agents to read `result.baseRef` which would be undefined. - The base-ref fallback "after fetching" is unreachable via the `git` MCP tool (it blocks `fetch` per AUTH_REQUIRED_REDIRECT). Now names `git_fetch` explicitly. - Boundary-tag wrapping for the user request had no escape rule for input that contains the literal close marker, and no fallback for an empty request. Both are now documented with a nonce-suffix mitigation. - PR reference updated #555 → #557 (the active PR for the multi-lens review-mode-orchestrator branch; #555 was closed after the rebase). Minor fixes: - Retry predicate tightened: "errors out (tool error) or returns an empty body", not "returns nothing usable" (which is unfalsifiable and lets an orchestrator declare any output not-usable to skip review). - Subagent read-only constraints rephrased as prescriptive ("MUST NOT call") rather than descriptive ("you have only"), since on inheriting runtimes the subagent does in fact have access to write tools and the constraint is prompt-only. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal round 2: tighten Build prompt edge cases (workflow_dispatch, base-ref, footer-strip, skip marker) Cross-lens findings from holistic + user-journey + research-validated lenses: - workflow_dispatch + empty diff: report_progress silently no-ops when there's no parent issue/PR. Now also call set_output with a "no-op" summary so the user gets surfacable feedback. - base-ref resolution: clarified `base` from checkout_pr is a bare ref name, added explicit `git remote show origin` path for repos whose primary is not `main` (master, trunk, etc.). - bare `git diff` description: tightened from "shows working tree" to "shows unstaged working-tree changes" — bare diff misses staged changes too, not just committed ones. - prompt-body stripping: explicitly call out the leading `> ` blockquote prefix (added by the *YOUR TASK* section formatting) and the entire Pullfrog footer block, not just one example link. - boundary-tag nonce: always-on now, not conditional on detecting a close marker. Cost is one random short string; failure mode (prompt injection if input contains literal close marker) is silent. - subagent-skip marker: structured `Self-review: SKIPPED (subagent error: ...)` on its own commit-message line, so the gap is greppable. Header comment also documents: - AddressReviews/Fix/Task asymmetry (deliberately deferred) - Subagent-runtime-fence deferred fix must explicitly deny Skill / agentic MCP tools, not just destructive tools (claude-code blocks recursive Task spawn but not alternative dispatch paths). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal round 3: targeted re-review of round-2 changes catches real regressions Round 2's "fixes" introduced two real bugs that round 3's targeted correctness re-review caught: CRITICAL (fixed): tier-3 base-ref resolution used `git remote show origin`, which requires network auth — the MCP `git` tool runs commands through plain spawn() without auth, so this hangs on private repos. Replaced with `git symbolic-ref refs/remotes/origin/HEAD` (local symref, no network), which actions/checkout populates. MAJOR (fixed): the eventInstructions fallback was incoherent — the agent has no separately-addressable eventInstructions field; whatever it received in *YOUR TASK* is its only input. Removed the misleading reference. MAJOR (fixed): per-line `> ` strip was ambiguous, could destructively flatten user-pasted markdown blockquotes. Now: "strip exactly one leading `> ` per line". MAJOR (fixed): tier-1 base-ref preferred bare `<base>` over `origin/<base>`, which fails on the rare alreadyOnBranch path in checkout_pr where the local ref isn't re-created. Now prefers `origin/<base>` (always populated post-fetch). MINOR (fixed): footer-strip anchor was `<sup>`/`<picture>`, both of which appear in legitimate user content (footnotes, etc.). Switched to the PULLFROG_DIVIDER sentinel which is purpose-built for this. MAJOR (acknowledged, partial fix): 4-hex nonce is theatrical security; bumped to 8 hex and explicitly noted it's a typo-guard, not a security boundary, and that the structural fix (separate task() argument) is the real solution. REJECTED (verified false positive): subagent claimed `set_output` is not registered for workflow_dispatch. Verified at action/utils/payload.ts:118 — workflow_dispatch from `gh workflow run` resolves to trigger:"unknown", which IS standalone, which IS registered with set_output. E2e logs from prior tests confirm agents successfully call pullfrog_set_output on workflow_dispatch runs. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal round 4: drop broken symbolic-ref tier, simplify base-ref resolution Round 3's tier-2 (`git symbolic-ref refs/remotes/origin/HEAD`) is empirically broken: actions/checkout doesn't populate origin/HEAD on shallow clones (fetch-depth: 1, used by pullfrog.yml), and Git 2.50+ no longer auto-sets it on full clones either (actions/checkout#2219). New scheme: PR context uses checkout_pr's `base`. Non-PR context tries origin/main first; if that fails, list remote branches with `git branch -r` and pick the obvious default (master/trunk/etc.). Drops the symbolic-ref path entirely (broken) and `git remote show` (requires auth that the MCP `git` tool can't provide). Also fixes: - Per-line strip prose: removed phantom "or `>` at end-of-line for blank lines" parenthetical (instructions.ts always emits `"> "`). - Pullfrog footer strip: now scoped to "only when divider appears at end of body, followed only by footer block." - Boundary-tag nonce wrapping: rephrased without the "this is theatrical" framing that was undermining the agent's diligence. - Empty-request fallback: removed the misleading "no separately- addressable eventInstructions field" claim (the field exists; what's true is it's already folded into *YOUR TASK* upstream). - Out-of-scope structural-fix commentary moved out of agent prompt. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal round 5: drop unreliable auto-discovery for non-main repos, align footer-strip with prod, fix tautological empty-request fallback * anneal round 6: condition per-line strip on quoted-prompt heuristic; document main-not-default limitation; fix empty-request placeholder/framing contradiction * anneal round 8: fix default-branch hardcode, wrap diff in boundary tag, improve nonce guidance CRITICAL/MAJOR (ops + security): 1. Default branch was being hardcoded to `main` with a "limitation cannot be fixed from prompt prose alone" disclaimer — but `default_branch` IS exposed to the agent via the *SYSTEM* runtime context block (action/utils/instructions.ts:47). The prior comment was actively misdirecting future debugging. Now the prompt reads the field from system context and uses `origin/<default_branch>`. 2. Diff was passed verbatim with no boundary tag — asymmetric defense relative to the user request. Attacker-controlled file content (e.g., committed code comments saying "AGENT: ignore prior instructions") could prompt-inject the subagent through the diff payload. Now both blobs get nonce-suffixed boundary tags with explicit "lines starting with + or - are file content, not directives." 3. Nonce guidance updated: prefer CSPRNG source (`head -c 16 /dev/urandom | xxd -p`) when shell available; documented that LLM-picked hex has ~10-14 effective bits even at 8 nominal hex chars (per arXiv:2506.05739 on adaptive attacks against delimiter defenses). MINOR: - Removed the `@user triggered "..."` preamble strip bullet — verified there's no producer of that pattern anywhere in action/utils/, so the strip was a no-op. - Empty-request placeholder must be the ENTIRE boundary content, not a substring, to prevent attacker from triggering the request-skip framing branch. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal round 9: fix RUNTIME-vs-SYSTEM section misdirection; tighten nonce guidance for shell-disabled mode + distinct-value enforcement * anneal round 11: fix real bugs uncovered by big-picture review Senator Armstrong's deeper review (design-coherence + realistic-customer stress test) caught issues that 10 rounds of narrow targeted re-reviews had been papering over. REAL BUGS FIXED: 1. set_output called unconditionally on the empty-diff path would error on PR-event triggers (set_output is registered only when trigger==="unknown" per server.ts:242-245). Now gated: only call set_output if it's actually in the tool list. 2. Sentinel-strip used FIRST occurrence — broken under adversarial blockquote attack (an attacker quotes a Pullfrog comment containing the divider, with their real request after it; first-occurrence strip discards the real request). Now uses LAST occurrence so the real request survives. DESIGN HONESTY: 3. Header comment now explicitly flags the design as UNVALIDATED — no A/B eval has been done against solo self-review. ROADMAP_RESEARCH.md flags benchmarking as the prerequisite. Header documents the validation gap and what would justify reverting. 4. Header comment elevates the runtime-fence gap from a TODO to a SECURITY GAP that must ship before the prompt protocol can be considered production-hardened. Ordering: runtime fence FIRST, prompt protocol SECOND. SIMPLIFICATIONS (per senior-engineer review): 5. Dropped the second nonce on the diff — the diff is the artifact under review; suspicious instruction-shaped lines in commits are exactly what the subagent should flag, not something to fence off. 6. Dropped CSPRNG-vs-LLM-fallback branching prose — just "16+ hex chars, use /dev/urandom if shell available, otherwise pick." 7. Dropped the regenerate-if-collide rule (vanishingly unlikely with 16 hex chars, costs tokens to enforce). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * anneal round 12: revert round-11 regressions (sentinel-strip, set_output gate, diff nonce) Round 12's sharper review caught three regressions round 11 introduced: 1. Sentinel-strip last-occurrence was strictly worse than first-occurrence for the common "user references a prior Pullfrog comment" case. The adversarial-quote scenario it was defending against is contrived (an attacker can put hostile payload anywhere; strip discipline doesn't change attack surface). Reverted to first-occurrence to align with canonical stripExistingFooter() and avoid silently swallowing user reference context. 2. set_output "gate" via "if it's in your tool list" relied on tool introspection that LLMs cannot reliably perform. Replaced with: just call report_progress; document the workflow_dispatch limitation as acceptable (job log is feedback-of-last-resort) rather than asking the agent to conditional-call a tool that may not exist. 3. Diff was de-nonced in round 11 on the assumption runtime fence ships first, but until that runtime fence lands the plain label is forgeable (committed file content can include "--- END DIFF ---" + injection). Restored nonce wrapping. The cost is one extra hex string; the benefit is real until runtime fence ships. Also added explicit caveat on the self-attested skip marker: the proper fix is MCP-layer dispatch-counting, not commit-message annotation. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * ruthless cut: revert Build self-review elaboration to compact form main already had subagent dispatch (4 compact lines). This branch added 70+ lines of elaboration — header warnings, base-ref dance, footer-strip rules, nonce- suffixed boundary tags, retry-once skip markers, delegation-discipline list — all predicated on a runtime fence that doesn't exist and validation that never ran. Senior-engineer review (round 11) explicitly recommended cutting; ROADMAP_RESEARCH flags A/B benchmarking as the prerequisite for this design. Net change vs main now matches what the user actually asked for: - drop the optional plan step (and its "follow the plan" / Notes references) - subagent receives the original user request alongside the diff, evaluated against base ref, with explicit no-further-dispatch constraint Everything else reverts to main's prose. ~10 lines net change instead of 70+. * anneal round 13: tighten self-review prompt inputs to runtime-resolvable values Two underspecified inputs flagged by parallel holistic + mechanics review: 1. "the original user request" is empty for non-@pullfrog-tagged auto-triggers (sync, check_suite, opened, etc.); only YOUR TASK is reliably present in the assembled prompt across all event types. Replace. 2. "base ref (PR base or repo default branch)" requires the agent to resolve and fetch the default branch on non-PR runs (origin/<default> typically not fetched). Drop the elaboration — bare git diff captures all changes at step-3 time since step 2 doesn't commit. Aligns with 3ed2c55a's ruthless-cut philosophy: less elaboration, not more. Verified in round 14: YOUR TASK is the literal section header in instructions.ts (buildTaskSection); bare git diff scope is correct. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * restore plan step to Build mode prompt The plan step was removed alongside the MCP-contract plan-required work, but the user only wanted it gone from the MCP contract, not from the prompt itself. Restores step 1 (plan), the "follow the plan" build sub-bullet, the trailing Notes section, and renumbers learningsStep back to 6. Made-with: Cursor * add pullfrog-reviewer named subagent; standardize review fence to non-mutative+non-recursive Defines a constrained `pullfrog-reviewer` named subagent for the Build mode self-review and /anneal lens dispatch, with a single source of truth in action/agents/reviewer.ts (allowed tools, denied mutating MCP tools, system prompt). Enforcement: - opencode: real fence via agent.pullfrog-reviewer block in buildSecurityConfig — denies edit/bash/task and globs each mutating pullfrog_* MCP tool to false. - claude-code: forward-looking only. Per-agent disallowedTools is upstream-broken (anthropics/claude-agent-sdk-typescript#172, open as of latest update Mar 2026 — subagent child processes still see and can call disallowed tools, including Task). The --agents JSON is defined anyway so the fence becomes real when upstream fixes #172; until then the prompt prose constraint is the actual fence. The PreToolUse hook workaround that does enforce is out of scope. Read-only MCP tools (get_*, list_*) intentionally remain enabled so the reviewer can pull PR/issue/check context without dispatching state changes. Both modes.ts Build self-review and the two anneal.md files now share the same "non-mutative + non-recursive" framing — file reads, grep, search, web search/fetch, read-only shell, and read-only MCP queries allowed; writes, state-changing MCP, and nested subagent dispatch denied. Resolves the previous inconsistency where /anneal allowed read-only shell and Build self-review banned all shell. Made-with: Cursor * Build self-review: pass build-phase failure summary to reviewer subagent Adds an instruction in step 4's dispatch: along with YOUR TASK and git diff, pass a tight plain-text summary of any lint/typecheck/test failures fixed during build (what broke, root cause, the fix) — or "no build-phase failures" if clean. Goal: let the reviewer check that fixes addressed root causes rather than suppressed symptoms (e.g., editing a test to make it pass instead of fixing the bug). Implemented as agent self-summarization rather than piping raw build output to avoid context flooding — typecheck/test output can be hundreds to thousands of lines per failure. The agent has the failure trail in its own conversation history and summarizes from memory; the reviewer sees a few lines per failure, not raw stderr. Caveat: this is a plausible-but-unvalidated quality improvement. The mechanical justification (signal already produced, currently not passed on) is real; "this catches more bugs" is a hypothesis that will need actual run data to confirm. Downside is bounded (reviewer gets slightly more context, no behavior change if the summary is empty or ignored). Made-with: Cursor * Build self-review: distill /anneal delegation + research discipline into dispatch instructions Lifts the codified learnings from /anneal's "Delegation discipline" and "Research discipline" sections into Build mode step 4. These rules are about how-to-prompt the reviewer (not about parallelism), so they transfer losslessly to single-agent dispatch and address bias modes the prior prompt was silent on: - Don't summarize what you implemented (biases toward shape-validation) - Don't curate a reading list (your curation is itself a lens) - Don't pre-shape output with severity/category (leaks hypotheses) - Don't defect-hunt in parallel (reintroduces the implementation bias the subagent is meant to mitigate) - For diffs touching third-party API contracts / SDK semantics / framework directives / DB engine specifics, instruct the reviewer to verify load-bearing claims via web search and quote URLs rather than trust training data Restructures step 4 from one paragraph into three (constraints, inputs, discipline) plus a final review-and-commit paragraph for readability. These are validated learnings from many anneal rounds, not theoretical best practices — they're the single substantive piece this branch was missing. Made-with: Cursor * pullfrog-reviewer: drop MCP deny-list, rely on prose constraint Per-PR-review feedback: hand-maintaining MUTATING_MCP_TOOLS against action/mcp/server.ts was fragile — a future mutating tool added to the MCP server without updating this list would silently grant write access to the reviewer. Inverting to an allowlist or adding a structural test both keep the drift problem. Drop the list and all per-agent runtime denies (claude disallowedTools, opencode tools/permission map). Strengthen REVIEWER_SYSTEM_PROMPT to spell out the categories of state-changing MCP tools by example and explicitly tell the model to apply the no-op-if-reverted invariant to tools added after the prompt was written — the rule is the invariant, not the enumeration. Keep the named subagent so the prompt is reliably injected. Update modes.ts and both anneal.md copies to drop the runtime-enforces-where-supported claim. Co-authored-by: Cursor <cursoragent@cursor.com> * pullfrog-reviewer: fix description to allow read-only shell The description field was overstating the constraint as 'must not shell', but the system prompt explicitly allows read-only commands like git diff, git log, cat, ls. Align description with the actual contract. Co-authored-by: Cursor <cursoragent@cursor.com> * restructure Review/IncrementalReview as multi-lens parallel-subagent orchestrators For someone else's PR, parallel lenses (correctness, security, research-validated claims, user-journey, etc.) provide breadth across angles that a single subagent can't carry coherently. The orchestrator does triage → parallel read-only subagent fan-out → aggregate → draft comments → submit. Lens count by risk: 1 lens for trivial PRs, 2-3 for typical, 4 for high-risk surfaces (billing, auth, migrations). This branch contains ONLY the Review/IncrementalReview multi-lens prompts. Build mode keeps its single-fresh-eyes-subagent shape (different problem — orchestrator just wrote the code; bias-mitigation comes from one subagent that doesn't share the implementation context). The Build changes ship in a separate PR (self-review-subagents → main). Pending validation against a real code-heavy PR before merge — e2e on a docs-only preview repo only exercised the trivial-1-lens path. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * Review/IncrementalReview: dispatch fan-out via reviewfrog named subagent The fan-out steps previously said "launch one read-only subagent per lens" without naming the subagent. That bypassed the only enforcement layer the named subagent provides: a baked-in system prompt that restates the non-mutative + non-recursive contract regardless of what the orchestrator sends. Both modes now dispatch via REVIEWER_AGENT_NAME (matching Build mode's self-review wiring) and restate the constraint inline so the rule is present twice. * rename pullfrog-reviewer → reviewfrog Mechanical rename of the named subagent. Constant names (REVIEWER_AGENT_NAME, REVIEWER_SYSTEM_PROMPT) and file paths (action/agents/reviewer.ts) stay as-is — only the agent identifier string and prose references in anneal.md and code comments change. * modes/anneal: trivial PRs skip review entirely; lens count is judgment, not table; allow subsystem lenses Three coupled changes to Review/IncrementalReview/Build self-review and the canonical /anneal command: 1. Trivial-skip: trivial diffs (single-line, formatting/comment-only, doc typo, low-risk dep bump, no behavior change) skip the fan-out / self-review entirely. Build mode skips its self-review subagent; Review submits a bare "Reviewed — no issues found." without dispatching lenses; IncrementalReview takes the existing non-substantive submit path. Tiebreaker on uncertainty: treat as non-trivial. 2. Drop prescriptive lens counts. Replaces "2-3 typical / 4 high-risk cap / 1 trivial" with judgment-based guidance: pick as many lenses as the target has distinct surfaces of risk worth investigating independently; one is sometimes enough; bias toward more (and toward follow-up rounds in /anneal) for high-stakes subsystems; 5+ is a smell that lenses are overlapping rather than covering distinct ground. 3. Subsystem lenses. Adds an explicit second flavor of lens — domain-scoped frames like "the auth lens", "the billing lens", "the schema-migration lens" — alongside the existing themed lenses (correctness, security, user-journey, etc.). Stack themed + subsystem freely. modes.ts and anneal.md (.cursor/ + .claude/, kept byte-identical) move together so the canonical pattern doc and the orchestrator prompt agree on the protocol. * add SessionLabeler so parallel subagent log lines are differentiable When the orchestrator dispatches multiple `reviewfrog` subagents in a single assistant turn (the parallel fan-out the multi-lens prompt now requires), their tool_use / tool_result / text events arrive on opencode's NDJSON stream tagged with distinct `sessionID`s but go through a single `[Pullfrog]` log prefix. Result: log readers can't attribute which lens issued which tool call, making CI logs unreadable for any review with 2+ lenses. SessionLabeler: - Binds the first-seen sessionID to "orchestrator" and subsequent new sessionIDs to FIFO-popped lens labels seeded from task tool_use inputs. - Derives labels from `lens: <name>` markers in the dispatch prompt, the Task `description` field, the `subagent_type`, or `subagent#N` fallback. - Keeps state local to a single runOpenCode invocation. Wiring: - opencode.ts: every event handler (init, message, text, tool_use, tool_result) now looks up the per-event label and prefixes log output via formatWithLabel(). Subagent finalOutput/token-reset paths gated on ORCHESTRATOR_LABEL so child sessions can't clobber parent state. - claude.ts: claude rolls subagent activity into a single tool_result block (no per-event session_id), so it gets a minimal "» dispatching subagent: <label>" log line on Task tool_use as the only attribution. - modes.ts (Review + IncrementalReview): orchestrator instructed to set the Task `description` to the lens name, since that's what the labeler reads when no explicit `lens:` marker is in the prompt. Tests: 18 unit tests covering label derivation, FIFO binding, interleaved sessions, fallback paths, and a realistic four-lens parallel fan-out simulation. Full action test suite stays green (400 passing). This is the pre-flight instrumentation that the multi-lens validation runs depend on — without it, post-hoc log analysis can't tell two subagents apart. * log subagent dispatch + finish at info level for per-lens visibility OpenCode's runtime currently encapsulates subagent execution inside the `task` tool — subagent-internal tool_use/tool_result events do not surface on the parent's NDJSON stream. The SessionLabeler I added in 0c4647f4 therefore can't actually differentiate concurrent subagent log lines (there are no concurrent log lines on the parent stream to differentiate). What CAN be observed on the parent stream is the dispatch and the result of each `task` tool call. This patch surfaces both at info level: » dispatching subagent: lens:security (subagent_type=reviewfrog) ... » subagent finished: lens:security (15.3s, status=completed) — ... Without this, a 4-lens parallel fan-out looks like 4 dispatches in close succession followed by a long quiet gap and then an aggregation turn — you can't see when each lens finished or how the durations overlapped. With it, parallel execution is visible from the timestamps on the "finished" lines. The dispatched label comes from SessionLabeler.recordTaskDispatch (so both lines share the same lens identity). taskDispatchInfo maps callID to {label, startedAt} so the matching tool_result can compute duration and emit the finished line. Also added a defensive comment on the SessionLabeler instantiation documenting that the per-event session-prefix path is currently dormant in the opencode runtime, but kept in place so attribution flips on automatically if/when opencode begins streaming subagent sessions. * fix subagent-finished log: hybrid exact+FIFO callID matching opencode does not consistently surface a tool_result callID matching the originating tool_use callID for the `task` tool, so the previous exact-match-only finish line never fired. Now we: - Dual-index task dispatches by callID AND in a FIFO queue. - Track non-task callIDs so we can identify "unrecognised callID" results as likely-task-with-mismatched-id. - On tool_result, exact-match first; fall back to FIFO when the output looks like a subagent reply (>300 chars) and the callID is unknown. - Flush leftover dispatches at run end with an "(inferred at run-end)" suffix so the gap is visible if subagent results arrive entirely off the tool_result event path (e.g. inlined into the next assistant message). * fix subagent-finished log: move run-end flush to post-subprocess block Investigation on T3 + finish-log-validation runs revealed two real issues with my prior attempt: 1. The `result` event handler is dead — opencode never emits a `result`-typed event over its NDJSON stream, so the inferred-at-run-end flush I had placed there never fired. Move the flush to right after `runSubprocess` returns where it actually executes. 2. The FIFO heuristic was too strict — the >300-char output check excluded short or empty outputs that opencode's `task` tool_result appears to carry (the subagent's full reply seems to arrive via a separate channel, not the result event itself). Drop the size check; rely solely on `knownNonTaskCallIDs` to keep genuinely-non-task tool_results from popping a pending task. Net effect: every `task` tool dispatch gets a matching `» subagent finished` line in the logs, either from the FIFO fallback during the run or from the run-end flush as a backstop. * modes/anneal: anchor lens calibration in worked examples The prior trivial-skip definition ("single-line fix, formatting-only, …") was anchored on diff size, but real-world risk is anchored on diff *shape*: a 5000-line lockfile regen IS trivial, and a 1-line SQL operator flip in a billing path is NOT. The prior lens-count guidance ("there's no fixed count, bias toward more for high-stakes subsystems") gave the agent no concrete shapes to anchor against, so runs varied between under-pick (4 generic lenses on a billing PR) and over-pick (5 overlapping themed lenses on a refactor). This commit hardens both: - Trivial definition gets explicit "looks trivial but isn't" anti-patterns: SQL operator flips, money/tax/timeout constants, feature-flag defaults, comparison operator changes, semantic 1-liners buried in whitespace, public-API renames, new direct deps. Skip lists get explicit "size doesn't matter" calibration for lockfile regens and mechanical renames. - Lens count gets a worked-example ladder: 1 lens (refactor / new test file / isolated fix), 2-3 lenses (typical features), 4-5 lenses (high-stakes subsystem touches), 6+ is a smell. - Subsystem lenses get an explicit recommendation to lead over generic themed equivalents for high-stakes domains, with the reasoning: domain framing primes the subagent for domain-specific failure modes (double-charges, refund races, dispute flows) the generic lens misses. Mirrored byte-identical into both anneal.md copies; modes.ts updates all three review surfaces (Build self-review, Review triage, IncrementalReview triage). * fix harness false-failure when Review submits without todowrite Review and IncrementalReview prompts explicitly forbid calling report_progress (the review IS the durable record). The post-run harness in action/utils/run.ts errors with "agent completed without reporting progress" when toolState.wasUpdated is false at exit. Until now, the only path that set wasUpdated for these modes was the todoTracker's debounced publish — which only fires if the agent happens to call todowrite during the run. Adversarial run on PR #16 (misleading-trivial billing tweak) hit exactly this case: agent went straight from triage → fan-out → review submission with no todowrite calls, and the harness reported failure even though the substantive review was successfully submitted with two inline comments. Fix: create_pull_request_review now marks wasUpdated=true (and finalSummaryWritten=true) on every terminal path — successful submit, empty-content skip, and all-comments-dropped skip. Submitting a review is unambiguously a "done" signal in these modes. Found via adversarial testing of the multi-lens orchestrator on a 1-line tax constant change. Logged in /tmp/pullfrog-validation/v3/. * fix harness false-failure when Review submits without todowrite (correctly) Replaces the prior fix (acc2bd65) which set wasUpdated=true inside create_pull_request_review. That approach worked for the harness check but broke the orphan-comment cleanup: with wasUpdated=true and finalSummaryWritten=true, the (!wasUpdated || trackerWasLastWriter) condition in main.ts evaluated false and the "Leaping into action" progress comment was left behind on every Review run — the exact behavior the cleanup logic was designed to prevent (see plans/review_progress_comment_cleanup_b0120f6c.plan.md). Correct fix: change the harness check in action/utils/run.ts to recognize a submitted PR review as an alternate completion signal alongside wasUpdated. wasUpdated stays false on purpose so cleanup deletes the orphan, but the run no longer false-fails when the agent followed the Review-mode contract (submit a review, never call report_progress). The bug was discovered during adversarial testing of PR #16 (misleading-trivial billing tweak) where the agent went straight from triage → fan-out → review submission without using todowrite, causing the harness to error even though the substantive review (a CAUTION blocking review with two inline comments catching a 10x tax cut) was successfully posted. * fix harness false-failure for Review modes (mode-based carve-out) Replaces the prior carve-out (4c0f69aa) which gated on toolState.review.id. That worked for runs where the review tool actually populated the toolState (validation-2 succeeded), but failed for runs that took a slightly different path where the assignment didn't propagate visibly to handleAgentResult — even when the review verifiably posted to GitHub. Found this empirically: PR #19 (pure mechanical rename across 20 files) opened with the prior fix in place, the agent picked exactly one impact lens (correct calibration!), confirmed no stale references, submitted "Reviewed — no issues found." successfully (visible in GitHub API), and the harness STILL errored with "agent completed without reporting progress." Same SHA, same branch, same code as validation-2 which passed. The toolState.review.id check turns out not to be reliably visible from the run.ts handler in all paths. Better fix: gate on toolState.selectedMode. Review and IncrementalReview modes are designed to never call report_progress (the review is the durable record, and IncrementalReview's non-substantive path produces no artifact at all by design). The harness completion check makes no sense for these modes — skip it entirely. The agent's clean subprocess exit is the completion signal. This also handles edge cases the previous fix missed: IncrementalReview's non-substantive path (no review submitted by design) and any future Review-flow shape that doesn't end at create_pull_request_review. * ci: trigger Test run to validate models-live timeout/concurrency changes * ci: prune passthrough models from live smoke matrix openrouter/* aliases and keyed opencode/* aliases are routing-layer wrappers around models we already smoke-test directly. running every passthrough burns CI minutes (~30 min/run) without catching anything the direct smoke doesn't — slug drift is already covered by the models-catalog job. keep one canary per routing layer (openrouter/claude-sonnet, opencode/claude-sonnet) to validate auth + tool-call translation. free opencode models stay in the matrix since they're unique to the provider. INCLUDE_ALL_PASSTHROUGHS=1 bypasses the prune for full validation. matrix size: 37 → 20 jobs. * fix isRateLimited false-positive on UUIDs/timestamps containing 429 The bare "429" substring pattern was matching MCP session IDs (e.g. `...-4429-...`) and microsecond timestamps in agent stdout, sending transient failures down the 60s rate-limit retry path. With the new 4-minute per-step CI timeout, that backoff plus a slow retry pushed the step past its budget and timed out. Switch to regex patterns and gate the numeric code on `\b429\b` so word boundaries prevent the substring false-match. Verified locally that the UUID `97287d2f-ae1d-4429-8627-73e2454e80ca` and timestamp `02:04:50.9429654` no longer match while real `HTTP 429` / `"status":429` strings still do. --------- Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: Colin McDonnell <colinmcd94@gmail.com> Co-authored-by: pullfrog[bot] <226033991+pullfrog[bot]@users.noreply.github.com>	2026-05-04 19:13:51 +00:00
Colin McDonnell	4b3c5ca905	rename agent key to opencode and add skill invocation coverage (#541) ... * rename agent key to opencode and add skill invocation coverage. add skill-invoke tests for claude and opencode with local play-based validation signals, update CI matrices, and include the current tracked refactors in this branch for review. Made-with: Cursor * exclude agent-specific skill-invoke tests from wrong agent in CI matrix * address review follow-up and preserve workflow run UI tweak. switch changed-agents ci coverage to exercise the opentoad implementation path while keeping the opencode expectation, and include the local workflow run client interaction updates requested on this branch. Made-with: Cursor * remove opentoad agent filename from runtime. rename the opencode harness implementation file from opentoad.ts to opencode.ts and update ci coverage input accordingly so action code no longer carries the old filename. Made-with: Cursor * ensure security prompt bypass is set on every test fixture. this keeps adversarial and permissions harnesses from being blocked by the default prompt-injection refusal path during CI security tests. Made-with: Cursor --------- Co-authored-by: pullfrog[bot] <226033991+pullfrog[bot]@users.noreply.github.com>	2026-04-15 19:38:36 +00:00
Colin McDonnell	0055aef618	feat: add Claude Code agent for Anthropic model users (#502) ... * feat: add Claude Code agent for Anthropic model users Re-adds Claude Code support (removed in #478) so users with Anthropic API keys or Claude Code OAuth tokens can use their Claude subscriptions directly. When an Anthropic model is selected and Claude Code credentials are available, the system auto-selects the Claude agent instead of OpenCode. The harness mirrors opentoad's security model: native Bash blocked via --disallowedTools, MCP ShellTool for restricted shell, ASKPASS for git auth. Includes NDJSON streaming, provider error detection, cache/cost tracking, browser skill, and todo progress tracking. Key changes: - action/agents/claude.ts: full Claude Code harness - action/utils/agent.ts: auto-select Claude for anthropic/* models - action/utils/providerErrors.ts: extracted shared provider error detection - action/utils/skills.ts: extracted shared skill installation (agent-aware) - action/models.ts: add CLAUDE_CODE_OAUTH_TOKEN to anthropic envVars - action/utils/docker.ts: add CLAUDE_CODE_OAUTH_TOKEN to test env allowlist - CI: add claude to test matrix, pass CLAUDE_CODE_OAUTH_TOKEN secret Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: remove unused toolId variable, fix apiKeys test env cleanup The apiKeys test cleanup stripped *_API_KEY vars but missed CLAUDE_CODE_OAUTH_TOKEN which doesn't match that pattern. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: strip provider prefix from PULLFROG_MODEL in Claude agent the env override path was returning the raw value (e.g. "anthropic/claude-sonnet-4-5") without stripping the provider prefix, causing the Claude CLI to receive an invalid model ID. Made-with: Cursor * fix: remove dead cliPath field, add CLAUDE_CODE_OAUTH_TOKEN to workflows remove unused cliPath from Claude agent RunParams, and pass CLAUDE_CODE_OAUTH_TOKEN through all pullfrog.yml workflow templates so users with Claude Pro/Team subscriptions can use their membership. Made-with: Cursor * fix: block Bash subagent in Claude Code disallowedTools Made-with: Cursor * chore: update model snapshot (opencode/openrouter latest → qwen3.6-plus-free) Made-with: Cursor --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>	2026-03-31 03:29:54 +00:00
Colin McDonnell	3ff11f97eb	replace deprecated opencode/mimo-v2-flash-free with mimo-v2-pro-free, update model snapshot ... Made-with: Cursor	2026-03-21 16:31:45 +00:00
Colin McDonnell	30d68e53a7	fix: skip API key validation for free opencode models ... free models (big-pickle, gpt-5-nano, etc.) define envVars: [] and isFree: true but validateAgentApiKey always required at least one provider key. now the validation is model-aware: free models bypass the check, keyed models validate their specific vars, and auto-select still requires at least one known key. closes #483 Made-with: Cursor	2026-03-16 20:48:59 +00:00