opencode v2: harness adapted to opencode-ai 1.15+ SDK-v2 / Effect-ts CLI rewrite (#767)

* opencode v2: harness adapted to opencode-ai 1.15+ SDK-v2 / Effect-ts CLI rewrite Bumps `opencode-ai` from `1.1.56` → `1.15.1` and ports the harness to the v2 NDJSON event contract. The legacy `opencode.ts` is kept as reference; `opencode_v2.ts` is the active runner via `agents/index.ts`. Why: `1.1.56` doesn't echo Gemini `thought_signature` back through the MCP tool-call serializer, so direct-Google reviews 400 on the 3rd-ish tool call. The fix only exists in the `1.14.x`+ line, which also ships the SDK-v2 / Effect-ts CLI rewrite — taking the rewrite is mandatory. Also unblocks the Codex ChatGPT-subscription auth path. Surface area: - drop `init` / `message` / `result` / `tool_result` event types and handlers (no longer emitted at v1.14+ per upstream `cli/cmd/run.ts:588-601`). - `tool_use` is now a single event covering both `state.status: "completed"` and `"error"`. duration / subagent-finish bookkeeping moves from the v1 `tool_result` handler into the consolidated `tool_use` handler. - new `reasoning` event handler — gated on `--thinking`, surfaces Gemini-3 / OpenAI / Anthropic thinking blocks. `--thinking` added to `baseArgs`. - drop `pendingTaskDispatches` FIFO + `knownNonTaskCallIDs` set: at v1.15 the `task` tool callID is stable across the whole `tool-input-* → tool-call → tool-result/tool-error` chain (`session/processor.ts:282-330`). exact-match map is sufficient. - drop `experimental.batch_tool: true` from injected config — declared but inert at v1.15. re-add once upstream wires it back. - bin path: `bin/opencode` → `bin/opencode.exe` (postinstall renames the platform-specific binary into `opencode.exe` for every OS now). Validated locally: - `pnpm test` 610/610 ✓ - `pnpm play --raw` end-to-end with Anthropic via OpenRouter ✓ - `pnpm play --raw` with `google/gemini-3.1-pro-preview`: 6 tool calls, multiple reasoning blocks visible, `set_output` propagates, exit 0 ✓ (this is the headline `thought_signature` fix) - runtest opencode: smoke ✓, restricted ✓, nobash ✓, token-exfil ✓ - runtest opencode: skill-invoke and mcpmerge fail (model-behavior drift on the new system prompt; wiring confirmed intact via direct repro showing both `robinMCP` and `pullfrog` MCP tools exposed). Tracked for follow-up; does not gate the migration. Plugin (`opencodePlugin.ts`) and skill discovery paths are unchanged at v1.15 — verified upstream and reused as-is. Bus subscription via `bus.subscribeAll()` and the `event` hook still fan out every payload. * model-smoke: bump opencode bin path to opencode.exe (v1.14+ rename) The v1.14+ postinstall.mjs renames the platform-specific binary to `bin/opencode.exe` for every OS (incl. linux/darwin), not just Windows. Mirrors the fix in action/agents/opencode_v2.ts. * opencode v2: set PWD env explicitly to fix skill / project-config discovery Root cause for skill-invoke + mcpmerge harness regressions: opencode-ai 1.15 reads `process.env.PWD` first (with `process.cwd()` as fallback) when resolving the SDK client's `directory` parameter — see upstream `cli/cmd/run.ts:282`: const root = Filesystem.resolve(process.env.PWD ?? process.cwd()) We pass `cwd: repoDir` to spawn, but the child inherits the harness's PWD via `...process.env`. Under `pnpm runtest` (and `pnpm play`) PWD is the `action/` directory, not the cloned test repo. Result: opencode creates two instances per session — one at `process.cwd()` (correct) and one at `PWD` (wrong) — and the agent's session runs in the PWD-derived one, which can't see the project's `.opencode/skills/` or `.claude/skills/`. Empirically traced via the full opencode stderr trace under the runtest harness: `service=skill count=3 init` (no `pullfrog-skill-check`) plus a second `service=default directory=<harness-pwd> creating instance` line per run. With `PWD=repoDir` set explicitly, `count=4 init` includes the test skill, the agent reaches for `skill({"name":"pullfrog-skill-check"})` exactly as the validator expects, and mcpmerge's `robinMCP_get_test_value` becomes accessible too. Validated locally: skill-invoke-opencode ✓, mcpmerge-opencode ✓, smoke ✓, restricted ✓, nobash ✓, token-exfil ✓ (flaked once on a model-narration match, passes on retry; unrelated to PWD). * opencode v2: drop ThinkingTimer; use opencode's reasoning.part.time directly opencode-ai 1.15 emits `reasoning` parts with `time.start` / `time.end` on terminal state (`cli/cmd/run.ts:671`), giving us a precise per-block "thought for X s" duration straight from the runtime. The v1 ThinkingTimer heuristic — measuring wall-clock between markToolResult and the next markToolCall — was an approximation when no native source existed; with v2 it's redundant and noisy (it would log alongside the real reasoning event, and conflated network latency with model thinking). Removed: `ThinkingTimer` import, `thinkingTimers` Map, `timerFor()` helper, both `markToolCall` / `markToolResult` call sites in `tool_use`. The `reasoning` handler now reads `part.time.start/end` directly and prefixes the visible preview with `(X.Ys)`. Output before: `» thinking: <preview>` + `» thought for 4.0s` (separate) Output now: `» thinking (4.0s): <preview>` (one line, sourced) For models that don't emit reasoning (Sonnet without extended thinking, GPT-4o, etc.), there's just no thinking line — which matches reality better than the gap-heuristic, which would fire on any pause >3s including provider-side latency that wasn't actual model reasoning. Validated locally: skill-invoke ✓, mcpmerge ✓, smoke ✓, Gemini play shows `» thinking (4.0s)` and `» thinking (0.8s)` from real durations. * claude.ts: same PWD fix as opencode v2; entryPost: refresh stale comment claude-code 2.1.x reads `process.env.PWD` and registers it as a "session" additional-working-directory when it differs from `process.cwd()` (per the bundled cli.js: `let H = process.env.PWD; if (H && H !== Y7() && ...) j.set(H, { path: H, source: "session" })`). Without overriding PWD on the spawn env, claude inherits the harness's PWD via `...process.env` — under `pnpm runtest` / `pnpm play` that's `action/`, not the cloned test repo — and adds the wrong dir to the agent's allowed working set. Symmetric to the opencode v2 fix in 52337f9. Pre-empts the same class of "agent's session sees the wrong cwd" failures on the claude side. Also refresh the stale `action/agents/opencode.ts` reference in entryPost.ts to point at opencode_v2.ts (the active runner), with the v1 file noted as kept-for-reference. * opencode: extract shared helpers into opencodeShared.ts; v2 cleanup Code-quality pass on the v2 work: 1. New `agents/opencodeShared.ts` (144 lines) for genuinely-shared helpers between v1 and v2: - `OpenCodeConfig` type - `geminiHighThinkingOverrides()` (registry-driven Gemini thinking pin) - `buildReviewerAgentConfig()` (reviewfrog config builder, was in v1 and re-imported by v2 via a back-reference) - `installOpencodeCli({ binPath })` (parameterized — v1 passes `bin/opencode`, v2 passes `bin/opencode.exe` via a per-version `installCli` lambda; matches each pinned version's npm shape) - `autoSelectModel()` + `getOpenCodeModels()` model-registry fallback v2 drops the `import { ... } from "./opencode.ts"` back-reference; v1 keeps a one-line `export { geminiHighThinkingOverrides }` re-export so `opencode.test.ts` keeps working unchanged. Once v1 is retired (post burn-in) opencodeShared collapses back into v2. 2. `opencode_v2.ts` cleanup: - drop dead state (`currentStepId`, `stepHistory` were write-only — their reader was the v1 `tool_result` handler we deleted) - hoist `state` in `tool_use` handler; replace nested-ternary payload extraction with a `terminalPayload(state)` helper - extract `formatPartDuration(time)` for the reasoning-block "(X.Ys)" suffix - tighten `OpenCodeBusEnvelopeEvent` type to include `tool` / `callID` fields directly, drop the `partWithToolFields` cast - trim docblocks per AGENTS.md "≤ 2-3 lines per code line": reasoning handler, tool_use handler, bus envelope handler all shortened - `step_start` becomes an explicit `() => {}` no-op so the dispatcher doesn't log "unhandled event" for every step 3. `subagentRegistration.test.ts` retargeted at the new file split — reads opencodeShared.ts for the buildReviewerAgentConfig assertions and opencode_v2.ts for the orchestrator-model wire-through. Net: -306 source lines (1339+1130 → 1228+1031+144). Tests + lint + format + typecheck all green; skill-invoke-opencode ✓ and smoke ✓ verified against the refactored v2 runtime. * opencode v2: address PR review feedback Three fixes from the inline review threads on #767: 1. Activity-diagnostic ordering bug (Copilot review at L705): the chunk- level `markActivity()` resets the module-level idle counter, so the per-event `getIdleMs()` sample inside the dispatch loop was always ~0ms — the "no activity for Xs" diagnostic never fired. Replaced with a runner-local `lastEventAt` so we measure real event-to-event silence instead of chunk-arrival latency. Drop the unused `getIdleMs` import. 2. TDZ-defensive hoist (Pullfrog review nit): `agentErrorEvent`, `lastProviderError`, and `recentStderr` are closed over by the `handlers` const but were declared after it. No current bug because handlers only fire inside the awaited `spawn()`, but a future refactor that triggers a handler synchronously during setup would surface a TDZ. Hoisted above `handlers`. 3. `step_finish.part.tokens.reasoning` follow-up (Pullfrog review at L566): leave a `TODO` comment marking the gap until `AgentUsage` grows a `reasoningTokens` field — separate PR with schema work. Cost totals stay correct because `part.cost` is summed independently. Other thread states for the record: - Copilot L63 (geminiHighThinkingOverrides import from legacy): already fixed by the opencodeShared.ts extraction in 83a7cab. - Copilot L672 (ThinkingTimer over-reports on terminal events): already fixed by dropping ThinkingTimer in a1e536b — we use opencode's own `reasoning.part.time.{start,end}` for thinking durations now. - Pullfrog L642 (onToolUse double-fire on subagent dispatch): re-checked the bus-envelope flow; the plugin filters orchestrator events except for status=running task dispatches, and bus-envelope returns before calling handlers.tool_use on those. No double-fire under current code. Validated: 610/610 unit tests, lint + format + typecheck clean, skill-invoke-opencode ✓. * DX: flip pnpm play / pnpm runtest to docker-by-default Restores the script shape wiki/docker.md has documented since the docker rewrite (#750). PR #756 inadvertently reverted action/package.json's gha/play/runtest scripts to host-only and dropped the :local variants; the wiki kept the new shape, so docs and reality drifted. The OpenCode-v2 migration agent ran `pnpm play --raw …` host-side throughout because the host entry was the only thing that existed. scripts (root → action): - pnpm play → pnpm -C action gha play.ts (docker, default) - pnpm play:local → pnpm -C action play:local (host) - pnpm runtest → pnpm -C action gha test/run.ts (docker, default) - pnpm runtest:local → pnpm -C action runtest:local (host) - pnpm gha is restored in action/package.json (re-adds `node gha.ts`) action/package.json deliberately ships only the :local variants — bare `pnpm -C action play` now errors instead of silently bypassing docker. This is a tradeoff per the user prompt's "consider whether NAMES should change" hint: the explicit error is worth the small CI churn. CI workflows: `.github/workflows/test.yml` and `action/.github/workflows/test.yml` flipped from `pnpm runtest …` to `pnpm runtest:local …`. Semantics unchanged — they still execute `node test/run.ts` directly on the GHA Linux runner; nesting docker on GHA is unnecessary overhead. Only the script name changed to match the new package.json. Webhook tester: the existing root `pnpm play` was actually a webhook handler smoke harness (root play.ts), unrelated to the action runtime. Renamed root play.ts → webhook.ts and exposed it as `pnpm webhook` so `pnpm play` can carry the docker-by-default action shortcut without collision. README updated. File headers updated: - action/play.ts: invocation block now points at `pnpm play` / `pnpm play:local` - action/test/run.ts: same - action/gha.ts: usage block calls out the new shortcut wrappers AGENTS.md: extended the existing "local sanity checks of action tool logic" rule with the play / play:local / runtest / runtest:local selection guidance and the `cd action; pnpm play` footgun note. wiki/docker.md unchanged — already described the now-real shape. * test/crossagent: add codex-auth smoke Pins openai/gpt-5.5 (in opencode's Codex ALLOWED_MODELS) and runs the full opencode harness against the env-provided CODEX_AUTH_JSON. Verifies: - installCodexAuth() materializes auth.json under the test HOME - opencode routes openai requests through ChatGPT subscription auth (no OPENAI_API_KEY in env, AT path forced via expires: 0) - the refresh chain advances during the run (refresh_token rotates) - detectCodexRefresh() would surface the rotation to entryPost.ts The post-hook write-back fetch isn't reachable from `pnpm runtest` (it's a separate GHA `post:` step). The integration boundary that matters end-to-end is "did the on-disk auth.json change in a way detectCodexRefresh recognizes" — that's exactly what this test asserts. CI wiring (already committed in a1c1fd4f as part of the DX flip): - .github/workflows/test.yml: CODEX_AUTH_JSON via secrets in action-agents env block - action/.github/workflows/test.yml: same; codex-auth in the hardcoded test matrix with a claude exclude The provisioning step on the user's side is `gh secret set CODEX_AUTH_JSON --repo pullfrog/app < auth.json`. ci.test.ts: expectedAgentEnvVars now includes provider `managedCredentials` so the "env vars cover all provider API keys" invariant stays self-correcting as more managed credentials land. * docs(codex-auth): make storage requirement unmissable A previous reviewing agent on this branch came away thinking `CODEX_AUTH_JSON` could live in GitHub Actions secrets. It can't — `entryPost.ts` rewrites the rotated refresh token after every run, and GH Actions secrets are immutable at runtime, so any non-Pullfrog-Postgres storage breaks the refresh chain on the first rotation (~1h silent expiry). - wiki/codex-auth.md: prominent `[!IMPORTANT]` callout above the fold, with the words "GitHub Actions secrets DO NOT WORK" verbatim and an enumeration of broken alternatives. - action/utils/codexHome.ts + action/entryPost.ts: header comments now loudly contrast Pullfrog secret store vs GH Actions and explain the writeback constraint. - AGENTS.md: terse one-bullet rule next to the model-resolution rule so future agents don't repeat the mistake. - .github/workflows/test.yml + action/.github/workflows/test.yml: added a comment marking the existing `secrets.CODEX_AUTH_JSON` injection as a CI smoke-testing shortcut, not the canonical pattern. CI wiring itself unchanged per scope. * auth codex: auto-open device URL, drop --scope flag - detect `https://auth.openai.com/codex/device...` from codex CLI output and best-effort launch it in the user's default browser (open / xdg-open / cmd start, wslview fallback on linux). gated so we only open once per flow; failures are swallowed so manual copy-paste still works. - drop the `--scope` flag entirely. the device-code flow is fundamentally interactive (browser approval), so a "skip-the-prompt" flag for just one of the prompts was dead weight. collapses scope selection to "always prompt on org-owned, always account on user-owned". * rename gha→docker, flip play/runtest defaults to host the previous shape conflated "real GitHub Actions" with the local docker container that mocks it, and made the slow docker path the default for fast-iteration scripts. - `action/gha.ts` → `action/docker.ts` (banner, --doctor, --help, image tag `pullfrog-docker:*`, volume `pullfrog-docker-node-modules-*`, tmpdir, error messages) - `pnpm play` / `pnpm runtest` now default to host (fast iteration); `pnpm play:docker` / `pnpm runtest:docker` run inside the container - `pnpm gha` → `pnpm docker` (the container runner shortcut) - `pnpm webhook` → `pnpm play:webhook` (fits the play: namespace; the bare name implied a webhook server, which hookdeck-cli already is) - update docs (`wiki/{docker,action-tests,billing,adversarial,browser}.md`, `README.md`, `AGENTS.md`), CI workflows (`.github/workflows/test.yml`, `action/.github/workflows/test.yml`), and code headers (`action/{play,test/run,utils/runFixture}.ts`, `webhook.ts`, `action/test/coverage.ts`) `action/commands/gha.ts` keeps its name — it's the real GitHub Actions entry point for the `pullfrog gha` CLI command (not the docker mock). * fix(codex): route post-hook writeback through apiFetch + conditional skip Three threads addressing PR #767 followups. action/entryPost.ts: replace raw fetch() with apiFetch() so the PUT /api/runtime/secret call carries the x-vercel-protection-bypass header/query when targeting a preview deployment. raw fetch silently 401s against the Vercel SSO gate, so every preview-env Codex run was losing its rotated refresh token. production is unaffected (no SSO). action/test/crossagent/codexAuth.ts: gate the test on CODEX_AUTH_JSON via new TestRunnerOptions.skipIf hook. when the secret is absent (forks, contributors without it), runTestForAgent short-circuits to a passing-with-skipped ValidationResult before any agent spawn — so the matrix's fail-fast: true setting doesn't cascade-cancel siblings. CI on pullfrog/app and dev-local with .env both still run the test for real. printSingleValidation/printResults now render skipped entries distinctly. doc/comment drift: - docs/codex-auth.mdx, wiki/codex-auth.md: drop stale --scope flag mention (removed in 10be96db, scope is now always interactively prompted or implicit). - wiki/codex-auth.md: tighten Claude-defense wording — materialization is agent-gated (opencode/opencode_v2 harness), not model-gated; opencode runs with non-OpenAI models still materialize the file, it's just not read. - action/Dockerfile, action/docker-entrypoint.sh: pnpm gha / gha.ts → pnpm docker / docker.ts (renamed in a2a63929). - app/api/runtime/secret/route.ts: refer to the save-time scope prompt instead of the dropped --scope flag. * smoke: force ≥2 tool calls; document test-bar in wiki + AGENTS upgrade crossagent/smoke prompt to call pullfrog_git status before set_output. this exercises the 2nd model→agent round-trip across every providers-live flagship, catching bugs like the Gemini thought_signature echo that single-tool-call tests can't see. also adds the "bar for adding new LLM-driven tests" section to wiki/action-tests.md and an extension to the existing AGENTS.md no-tests rule pointing at it — prefer upgrading existing matrix entries over adding new ones. local: pnpm runtest smoke opencode passes against both anthropic/claude-sonnet-4-6 and google/gemini-pro. --------- Co-authored-by: Colin McDonnell <colinmcd94@M1chelle.local>
2026-05-20 04:05:16 +00:00
parent 4d1fd5ea1a
commit a0576a702a
30 changed files with 1658 additions and 312 deletions
@@ -949,9 +949,20 @@ export const claude = agent({
    // bedrock run; if the user has set the env var manually for some other
    // reason (e.g. always-Bedrock org policy), `...process.env` already
    // carries it through and we don't disturb it.
+    const repoDir = process.cwd();
+
+    // PWD must match the spawn cwd (see opencode_v2.ts for the analogous fix).
+    // claude-code 2.1.x reads `process.env.PWD` and registers it as a "session"
+    // additional-working-directory when it differs from `process.cwd()` (per
+    // the bundled cli.js — `let H=process.env.PWD; if(H && H !== Y7() && ...)
+    // j.set(H, {path: H, source: "session"})`). Inheriting harness PWD via
+    // `...process.env` ends up adding the wrong dir to the agent's allowed
+    // working set under `pnpm runtest` / `pnpm play`, which silently confuses
+    // path-relative tools.
    const env: Record<string, string | undefined> = {
      ...process.env,
      ...homeEnv,
+      PWD: repoDir,
    };
    if (isBedrockRoute) {
      env.CLAUDE_CODE_USE_BEDROCK = "1";
@@ -967,8 +978,6 @@ export const claude = agent({
      delete env.ANTHROPIC_API_KEY;
    }

-    const repoDir = process.cwd();
-
    log.info(`» effort: ${effort}`);
    log.debug(`» starting Pullfrog (Claude Code): node ${baseArgs.join(" ")}`);
    log.debug(`» working directory: ${repoDir}`);
@@ -1,5 +1,8 @@
 import { claude } from "./claude.ts";
-import { opencode } from "./opencode.ts";
+// v2 harness — adapted to opencode-ai >=1.14.x SDK-v2 / Effect-ts CLI rewrite.
+// The legacy v1 module (`./opencode.ts`) is kept around for reference + fast
+// revert; the active runner is the v2 module below.
+import { opencode } from "./opencode_v2.ts";
 import type { Agent } from "./shared.ts";

 export type { Agent, AgentUsage } from "./shared.ts";
@@ -11,13 +11,12 @@
 * the agent process itself gets full env (needs LLM API keys, PATH, etc.).
 * security is enforced at the tool layer, not the process layer.
 */
-import { execFileSync } from "node:child_process";
 import { mkdirSync, writeFileSync } from "node:fs";
 import { join } from "node:path";
 import { performance } from "node:perf_hooks";
 import * as core from "@actions/core";
 import { pullfrogMcpName } from "../external.ts";
-import { BEDROCK_MODEL_ID_ENV, modelAliases } from "../models.ts";
+import { BEDROCK_MODEL_ID_ENV } from "../models.ts";
 import type { ToolState } from "../toolState.ts";
 import {
  getIdleMs,
@@ -29,7 +28,6 @@ import {
 import { type AgentDiagnostic, formatAgentHangBody } from "../utils/agentHangReport.ts";
 import { formatJsonValue, log } from "../utils/cli.ts";
 import { installCodexAuth } from "../utils/codexHome.ts";
-import { installFromNpmTarball } from "../utils/install.ts";
 import { findProviderErrorMatch } from "../utils/providerErrors.ts";
 import { addSkill, installBundledSkills } from "../utils/skills.ts";
 import {
@@ -47,12 +45,19 @@ import {
  PULLFROG_OPENCODE_PLUGIN_FILENAME,
  PULLFROG_OPENCODE_PLUGIN_SOURCE,
 } from "./opencodePlugin.ts";
+import {
+  autoSelectModel,
+  buildReviewerAgentConfig,
+  geminiHighThinkingOverrides,
+  installOpencodeCli,
+  type OpenCodeConfig,
+} from "./opencodeShared.ts";
 import {
  buildLearningsReflectionPrompt,
  runPostRunRetryLoop,
  shouldRunReflection,
 } from "./postRun.ts";
-import { REVIEWER_AGENT_NAME, REVIEWER_SYSTEM_PROMPT } from "./reviewer.ts";
+import { REVIEWER_AGENT_NAME } from "./reviewer.ts";
 import { formatWithLabel, ORCHESTRATOR_LABEL, SessionLabeler } from "./sessionLabeler.ts";
 import {
  type AgentResult,
@@ -62,29 +67,13 @@ import {
  logTokenTable,
  MAX_STDERR_LINES,
 } from "./shared.ts";
-import { deriveSubagentModels } from "./subagentModels.ts";

-async function installOpencodeCli(): Promise<string> {
-  return await installFromNpmTarball({
-    packageName: "opencode-ai",
-    version: getDevDependencyVersion("opencode-ai"),
-    executablePath: "bin/opencode",
-    installDependencies: true,
-  });
-}
+// re-export for the existing test (`./opencode.test.ts`) — once v1 is
+// retired this module collapses and the test imports from opencodeShared.
+export { geminiHighThinkingOverrides } from "./opencodeShared.ts";

-// ── config ─────────────────────────────────────────────────────────────────────
-
-type OpenCodeConfig = {
-  mcp?: Record<string, unknown>;
-  permission?: Record<string, unknown>;
-  provider?: Record<string, unknown>;
-  agent?: Record<string, unknown>;
-  experimental?: Record<string, unknown>;
-  model?: string;
-  enabled_providers?: string[];
-  [key: string]: unknown;
-};
+// v1.4-era npm package shipped a per-platform binary directly at this path.
+const installCli = () => installOpencodeCli({ binPath: "bin/opencode" });

 // NOTE: OpenCode's per-call `max_tokens` defaults to 32_000. We previously
 // overrode this via `OPENCODE_EXPERIMENTAL_OUTPUT_TOKEN_MAX = 5000` in #616
@@ -106,22 +95,6 @@ type OpenCodeConfig = {
 // top-level `limit.output` config field has no read site (silently dropped
 // on merge in session/llm.ts), so the env var is the only working knob.

-/**
- * Build the `provider.google.models[id].options` map that pins every direct-Google
- * Gemini alias to `thinkingLevel: "high"`. Sourced from the model registry so
- * adding/renaming a Google alias in `action/models.ts` flows through automatically.
- */
-export function geminiHighThinkingOverrides(): Record<string, { options: object }> {
-  return Object.fromEntries(
-    modelAliases
-      .filter((a) => a.provider === "google")
-      .map((a) => [
-        a.resolve.replace(/^google\//, ""),
-        { options: { thinkingConfig: { thinkingLevel: "high" } } },
-      ])
-  );
-}
-
 function buildSecurityConfig(ctx: AgentRunContext, model: string | undefined): string {
  const config: OpenCodeConfig = {
    permission: {
@@ -171,90 +144,6 @@ function buildSecurityConfig(ctx: AgentRunContext, model: string | undefined): s
  return JSON.stringify(config);
 }

-/**
- * Read-only `reviewfrog` subagent for lens-based review.
- *
- * Non-mutative + non-recursive — enforced by the prose system prompt in
- * reviewer.ts.
- *
- * Per-subagent `model:` override is driven by the registry in
- * `action/models.ts` via each alias's `subagentModel` field — see
- * `deriveSubagentModels` for the reverse-lookup. Currently wired:
- * Anthropic opus → sonnet, OpenAI gpt-pro → gpt and gpt → gpt-5.4,
- * Google gemini-pro → gemini-flash. Other providers (xai, deepseek,
- * moonshot) and already-cheap tiers inherit (no override) — either the
- * absolute savings are too small to justify or there's no clean
- * cheaper-but-capable sibling.
- */
-function buildReviewerAgentConfig(orchestratorModel: string | undefined): Record<string, unknown> {
-  const overrides = deriveSubagentModels(orchestratorModel);
-  return {
-    [REVIEWER_AGENT_NAME]: {
-      description:
-        "Read-only review subagent for lens-based code review (correctness, security, billing-subsystem, etc.). " +
-        "Reads only — no writes, no state-changing shell or MCP calls, no nested subagent dispatch.",
-      mode: "subagent",
-      prompt: REVIEWER_SYSTEM_PROMPT,
-      ...(overrides.reviewer !== undefined ? { model: overrides.reviewer } : {}),
-    },
-  };
-}
-
-// ── model auto-select fallback ──────────────────────────────────────────────────
-//
-// steps 1–2 of model resolution (PULLFROG_MODEL env, slug resolution) are handled
-// by resolveModel() in utils/agent.ts before the agent runs. this fallback only
-// handles step 3: auto-select via `opencode models`.
-
-function getOpenCodeModels(cliPath: string): string[] {
-  try {
-    const output = execFileSync(cliPath, ["models"], {
-      encoding: "utf-8",
-      timeout: 30_000,
-      env: process.env,
-    });
-    return output
-      .split("\n")
-      .map((line) => line.trim())
-      .filter(Boolean);
-  } catch (error) {
-    log.debug(
-      `» failed to run \`opencode models\`: ${error instanceof Error ? error.message : String(error)}`
-    );
-    return [];
-  }
-}
-
-const AUTO_SELECT_WARNING =
-  "select a model explicitly in the Pullfrog console (https://pullfrog.com/console) to avoid this.";
-
-function autoSelectModel(cliPath: string): string | undefined {
-  const availableModels = getOpenCodeModels(cliPath);
-  const availableSet = new Set(availableModels);
-  if (availableSet.size > 0) {
-    log.debug(`» opencode models (${availableSet.size}): ${availableModels.join(", ")}`);
-    // skip hidden aliases (internal subagent-tier targets like opencode/gpt-5.4) —
-    // they should never surface as a user-facing orchestrator pick. mirrors the
-    // selectable-list filter in components/ModelSelector.tsx and action/commands/init.ts.
-    const match =
-      modelAliases.find((a) => !a.hidden && a.preferred && availableSet.has(a.resolve)) ??
-      modelAliases.find((a) => !a.hidden && availableSet.has(a.resolve));
-    if (match) {
-      log.info(
-        `» model: ${match.resolve} (auto-selected${match.preferred ? " — preferred" : ""} curated match)`
-      );
-      log.warning(`» model auto-selected. ${AUTO_SELECT_WARNING}`);
-      return match.resolve;
-    }
-    log.info(
-      `» opencode has ${availableSet.size} models but none match curated aliases — letting OpenCode auto-select`
-    );
-  }
-
-  log.warning(`» no model resolved. letting OpenCode auto-select. ${AUTO_SELECT_WARNING}`);
-  return undefined;
-}
-
 // ── NDJSON event types ─────────────────────────────────────────────────────────

 interface OpenCodeInitEvent {
@@ -1220,9 +1109,9 @@ async function runOpenCode(params: RunParams): Promise<AgentResult> {

 export const opencode = agent({
  name: "opencode",
-  install: installOpencodeCli,
+  install: installCli,
  run: async (ctx) => {
-    const cliPath = await installOpencodeCli();
+    const cliPath = await installCli();

    const rawModel = ctx.payload.proxyModel ?? ctx.resolvedModel ?? autoSelectModel(cliPath);

@@ -0,0 +1,144 @@
+// Shared helpers for the OpenCode agent harnesses (`./opencode.ts` v1 and
+// `./opencode_v2.ts` v2). Pure config / model-registry / install glue —
+// nothing here touches the NDJSON event loop, which differs between v1 and v2.
+//
+// Once v1 is deleted post-burn-in this module collapses back into v2; until
+// then it keeps both runners synchronized so a config drift can't make v1 a
+// silently-broken fallback.
+
+import { execFileSync } from "node:child_process";
+import { modelAliases } from "../models.ts";
+import { log } from "../utils/cli.ts";
+import { installFromNpmTarball } from "../utils/install.ts";
+import { getDevDependencyVersion } from "../utils/version.ts";
+import { REVIEWER_AGENT_NAME, REVIEWER_SYSTEM_PROMPT } from "./reviewer.ts";
+import { deriveSubagentModels } from "./subagentModels.ts";
+
+// ── config ─────────────────────────────────────────────────────────────────────
+
+export type OpenCodeConfig = {
+  mcp?: Record<string, unknown>;
+  permission?: Record<string, unknown>;
+  provider?: Record<string, unknown>;
+  agent?: Record<string, unknown>;
+  experimental?: Record<string, unknown>;
+  model?: string;
+  enabled_providers?: string[];
+  [key: string]: unknown;
+};
+
+/**
+ * Build the `provider.google.models[id].options` map that pins every direct-Google
+ * Gemini alias to `thinkingLevel: "high"`. Sourced from the model registry so
+ * adding/renaming a Google alias in `action/models.ts` flows through automatically.
+ */
+export function geminiHighThinkingOverrides(): Record<string, { options: object }> {
+  return Object.fromEntries(
+    modelAliases
+      .filter((a) => a.provider === "google")
+      .map((a) => [
+        a.resolve.replace(/^google\//, ""),
+        { options: { thinkingConfig: { thinkingLevel: "high" } } },
+      ])
+  );
+}
+
+/**
+ * Read-only `reviewfrog` subagent for lens-based review. Non-mutative +
+ * non-recursive — enforced by the system prompt in reviewer.ts.
+ *
+ * Per-subagent `model:` override is driven by the registry in
+ * `action/models.ts` via each alias's `subagentModel` field. Currently wired:
+ * Anthropic opus → sonnet, OpenAI gpt-pro → gpt and gpt → gpt-5.4, Google
+ * gemini-pro → gemini-flash. Other providers inherit (no override).
+ */
+export function buildReviewerAgentConfig(
+  orchestratorModel: string | undefined
+): Record<string, unknown> {
+  const overrides = deriveSubagentModels(orchestratorModel);
+  return {
+    [REVIEWER_AGENT_NAME]: {
+      description:
+        "Read-only review subagent for lens-based code review (correctness, security, billing-subsystem, etc.). " +
+        "Reads only — no writes, no state-changing shell or MCP calls, no nested subagent dispatch.",
+      mode: "subagent",
+      prompt: REVIEWER_SYSTEM_PROMPT,
+      ...(overrides.reviewer !== undefined ? { model: overrides.reviewer } : {}),
+    },
+  };
+}
+
+// ── install ────────────────────────────────────────────────────────────────────
+
+/**
+ * Install the opencode-ai npm tarball and return the path to the executable.
+ *
+ * The bin path differs by version: v1.4.x and earlier shipped `bin/opencode`;
+ * v1.14+ renames the platform-specific binary to `bin/opencode.exe` for every
+ * OS via the postinstall script. Callers pass the binPath that matches their
+ * pinned version so a v1↔v2 swap can't silently install the wrong file.
+ */
+export async function installOpencodeCli(params: { binPath: string }): Promise<string> {
+  return await installFromNpmTarball({
+    packageName: "opencode-ai",
+    version: getDevDependencyVersion("opencode-ai"),
+    executablePath: params.binPath,
+    installDependencies: true,
+  });
+}
+
+// ── model auto-select fallback ──────────────────────────────────────────────────
+//
+// steps 1–2 of model resolution (PULLFROG_MODEL env, slug resolution) happen
+// in resolveModel() in utils/agent.ts before the agent runs. this is step 3:
+// auto-select via `opencode models`.
+
+const AUTO_SELECT_WARNING =
+  "select a model explicitly in the Pullfrog console (https://pullfrog.com/console) to avoid this.";
+
+function getOpenCodeModels(cliPath: string): string[] {
+  try {
+    const output = execFileSync(cliPath, ["models"], {
+      encoding: "utf-8",
+      timeout: 30_000,
+      env: process.env,
+    });
+    return output
+      .split("\n")
+      .map((line) => line.trim())
+      .filter(Boolean);
+  } catch (error) {
+    log.debug(
+      `» failed to run \`opencode models\`: ${error instanceof Error ? error.message : String(error)}`
+    );
+    return [];
+  }
+}
+
+export function autoSelectModel(cliPath: string): string | undefined {
+  const availableModels = getOpenCodeModels(cliPath);
+  const availableSet = new Set(availableModels);
+  if (availableSet.size > 0) {
+    log.debug(`» opencode models (${availableSet.size}): ${availableModels.join(", ")}`);
+    // skip hidden aliases (internal subagent-tier targets like
+    // opencode/gpt-5.4) — they should never surface as a user-facing
+    // orchestrator pick. mirrors the selectable-list filter in
+    // components/ModelSelector.tsx and action/commands/init.ts.
+    const match =
+      modelAliases.find((a) => !a.hidden && a.preferred && availableSet.has(a.resolve)) ??
+      modelAliases.find((a) => !a.hidden && availableSet.has(a.resolve));
+    if (match) {
+      log.info(
+        `» model: ${match.resolve} (auto-selected${match.preferred ? " — preferred" : ""} curated match)`
+      );
+      log.warning(`» model auto-selected. ${AUTO_SELECT_WARNING}`);
+      return match.resolve;
+    }
+    log.info(
+      `» opencode has ${availableSet.size} models but none match curated aliases — letting OpenCode auto-select`
+    );
+  }
+
+  log.warning(`» no model resolved. letting OpenCode auto-select. ${AUTO_SELECT_WARNING}`);
+  return undefined;
+}
@@ -3,7 +3,8 @@ import { join } from "node:path";
 import { describe, expect, it } from "vitest";

 const claudeSource = readFileSync(join(__dirname, "claude.ts"), "utf-8");
-const opencodeSource = readFileSync(join(__dirname, "opencode.ts"), "utf-8");
+const opencodeSharedSource = readFileSync(join(__dirname, "opencodeShared.ts"), "utf-8");
+const opencodeV2Source = readFileSync(join(__dirname, "opencode_v2.ts"), "utf-8");

 /**
 * The Claude Code `--agents` JSON and OpenCode `agent` config block are the
@@ -25,16 +26,16 @@ describe("subagent registration source asserts", () => {
    });
  });

-  describe("opencode.ts buildReviewerAgentConfig", () => {
+  describe("opencodeShared.ts buildReviewerAgentConfig", () => {
    it("registers reviewfrog with mode: subagent", () => {
-      expect(opencodeSource).toMatch(/\[REVIEWER_AGENT_NAME\]:[^}]*mode:\s*"subagent"/s);
+      expect(opencodeSharedSource).toMatch(/\[REVIEWER_AGENT_NAME\]:[^}]*mode:\s*"subagent"/s);
    });
    it("uses deriveSubagentModels for the reviewer model override", () => {
-      expect(opencodeSource).toMatch(/deriveSubagentModels\(/);
-      expect(opencodeSource).toMatch(/overrides\.reviewer/);
+      expect(opencodeSharedSource).toMatch(/deriveSubagentModels\(/);
+      expect(opencodeSharedSource).toMatch(/overrides\.reviewer/);
    });
-    it("passes orchestrator model to buildReviewerAgentConfig", () => {
-      expect(opencodeSource).toMatch(/buildReviewerAgentConfig\(model\)/);
+    it("v2 runner passes orchestrator model to buildReviewerAgentConfig", () => {
+      expect(opencodeV2Source).toMatch(/buildReviewerAgentConfig\(model\)/);
    });
  });
 });