Agent & model refactor (#478)
* agent & model refactor with ASKPASS git auth, UI restructure, clerk v7 Made-with: Cursor * fix stale agent/effort refs, add tests for askpass + model resolution - reviewCleanup.ts: payload.agent -> payload.model, remove effort - selectMode.ts PlanEdit: remove delegation/subagent/effort references - pullfrog.yml.ts: update env vars (drop GOOGLE_API_KEY/CURSOR_API_KEY, add GOOGLE_GENERATIVE_AI_API_KEY/XAI_API_KEY/MOONSHOT_API_KEY/OPENCODE_API_KEY) - FlagsSettings/RepoInstructionsSection: remove stale effort/timeout copy - new: gitAuthServer.test.ts (10 tests — lifecycle, token delivery, tamper detection, script gen) - new: agent.test.ts (4 tests — default opentoad, AGENT_OVERRIDE, invalid override) - new: models.test.ts (19 tests — parseModel, resolution, registry invariants) - update models.dev snapshot Made-with: Cursor * fix changed-agents.sh to filter legacy agent files from CI matrix legacy agent files (claude.ts, codex.ts, etc.) are @ts-nocheck and not exported from index.ts. changed-agents.sh now reads index.ts imports to build the active agent set and treats changes to inactive files as non-agent changes (opentoad canary only). Made-with: Cursor * remove MCP file tools, old agent harnesses, and obsolete security tests ASKPASS-based git auth makes the old MCP file tool security layer unnecessary: - token never in subprocess env, so symlink/gitattributes/hook attacks can't exfiltrate it - agents now use native file tools (OpenCode builtin read/edit) deleted: - action/mcp/file.ts (file_read, file_write, file_edit, file_delete, list_directory) - action/mcp/index.ts (dead re-export) - agent harnesses: claude.ts, codex.ts, cursor.ts, gemini.ts, opencode.ts - opencode-runner.ts (inlined into opentoad.ts) - security tests that validated MCP file tool restrictions - commented-out three-step review flow (~300 lines) - sanitizeSchema/wrapSchema dead code from mcp/shared.ts - OPENCODE_MODEL_MINI/MAX env vars (effort-level model overrides removed) updated test prompts to use generic file ops instead of MCP tool names. restored pkg-json-scripts + requirements-txt-attack (test --ignore-scripts defense). Made-with: Cursor * bump actions/checkout v4 → v6 (node 24) node 20 actions deprecated june 2, 2026. Made-with: Cursor * temporarily disable fail-fast on agnostic tests to debug checkout@v6 Made-with: Cursor * re-enable fail-fast on agnostic tests Made-with: Cursor * fix test token mismatch: mint OIDC tokens scoped to target repo CI tests override GITHUB_REPOSITORY to pullfrog/test-repo but inherit the runner's GITHUB_TOKEN (scoped to pullfrog/app), causing 401s on every run-context fetch. Clear GITHUB_TOKEN in the test subprocess so ensureGitHubToken() mints a properly scoped token via OIDC. Also centralizes the default GITHUB_REPOSITORY in runAgentStreaming instead of repeating it in every test file, and fixes preview-cleanup to remove workers from all queues (not just name-matching ones). Made-with: Cursor * fix ensureGitHubToken to try OIDC when app credentials are absent ensureGitHubToken only attempted token minting when GITHUB_APP_ID and GITHUB_PRIVATE_KEY were set. In CI, OIDC is available but app creds aren't exposed — so the guard prevented minting entirely. Made-with: Cursor * dead code cleanup: remove remnants of deleted agents, file tools, effort system remove unused @anthropic-ai/claude-agent-sdk and @openai/codex-sdk deps, orphaned file-tool security tests, dead GEMINI_MODEL passthrough, stale opencode-runner wiki refs, deleted test file references, and MCP file tool docs. rename docs/effort → docs/models. fix vitest setup: move dotenv to globalSetup (runs once before forks instead of per-file, 19s → 200ms). Made-with: Cursor * address review feedback: remove dead code, update stale references - remove AGENT_OVERRIDE (only opentoad exists) - remove shellToolName plumbing (always restricted shell) - bump action version to 0.0.179 - remove CURSOR_API_KEY from all workflows/configs - remove OPENCODE_MODEL_MINI/MAX from workflows/docs - delete wiki/effort.md, rewrite docs/effort.mdx as "Models" - rewrite wiki/modes.md: orchestrator/subagent → single agent - simplify flag system: drop builtin flag extraction (debug, effort, timeout, agent), keep custom flag replacement only - reserve all legacy flag names to prevent custom flag conflicts Made-with: Cursor * regenerate lockfile after removing claude-agent-sdk and codex-sdk Made-with: Cursor * fix import ordering, add lockfile check to pre-push hook Made-with: Cursor * remove dead debug payload field, stale packageExtensions Made-with: Cursor * merge proc-sandbox and token-exfil into a single test proc-sandbox and token-exfil were duplicative — both tested that SANDBOX_TEST_TOKEN couldn't be exfiltrated. consolidated into token-exfil with shell:restricted (which actually exercises filterEnv) and the /proc attack vector hints from proc-sandbox. Made-with: Cursor * fix wiki adversarial.md to match actual tokenExfil validator Made-with: Cursor
This commit is contained in:
Colin McDonnell
committed by
pullfrog[bot]
pullfrog[bot]
parent
5bcfae990a
commit
6d25adfd1a
+3
-135
@@ -1,4 +1,4 @@
|
||||
import type { StandardJSONSchemaV1, StandardSchemaV1 } from "@standard-schema/spec";
|
||||
import type { StandardSchemaV1 } from "@standard-schema/spec";
|
||||
import { encode as toonEncode } from "@toon-format/toon";
|
||||
import type { FastMCP, Tool } from "fastmcp";
|
||||
import { formatJsonValue, log } from "../utils/cli.ts";
|
||||
@@ -61,141 +61,9 @@ export const execute = <T, R extends Record<string, any> | string>(
|
||||
return _fn;
|
||||
};
|
||||
|
||||
/**
|
||||
* Sanitize JSON schema to remove problematic fields that Gemini CLI/API can't handle
|
||||
* - Removes $schema field (causes "no schema with key or ref" errors)
|
||||
* - Converts $defs to definitions (draft-07 compatibility)
|
||||
* - Removes any draft-2020-12 specific features
|
||||
* - Converts any_of with enum values to direct STRING enum (Google API requirement)
|
||||
*/
|
||||
function sanitizeSchema(schema: any): any {
|
||||
if (!schema || typeof schema !== "object") {
|
||||
return schema;
|
||||
}
|
||||
|
||||
if (Array.isArray(schema)) {
|
||||
return schema.map(sanitizeSchema);
|
||||
}
|
||||
|
||||
// handle any_of with enum values - convert to direct STRING enum for Google API
|
||||
// Google API requires: {type: "string", enum: [...]} not {anyOf: [{enum: [...]}, {enum: [...]}]}
|
||||
if (schema.anyOf && Array.isArray(schema.anyOf) && schema.anyOf.length > 0) {
|
||||
const enumValues: string[] = [];
|
||||
let allAreEnumObjects = true;
|
||||
|
||||
for (const item of schema.anyOf) {
|
||||
if (item && typeof item === "object" && Array.isArray(item.enum)) {
|
||||
// collect enum values (only strings)
|
||||
const stringEnums = item.enum.filter((v: any) => typeof v === "string");
|
||||
if (stringEnums.length > 0) {
|
||||
enumValues.push(...stringEnums);
|
||||
} else {
|
||||
allAreEnumObjects = false;
|
||||
break;
|
||||
}
|
||||
} else {
|
||||
allAreEnumObjects = false;
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
// if all any_of items are enum objects with string values, convert to direct STRING enum
|
||||
if (allAreEnumObjects && enumValues.length > 0) {
|
||||
const uniqueEnums = [...new Set(enumValues)];
|
||||
// preserve other properties from the original schema (like description)
|
||||
const result: any = {
|
||||
type: "string",
|
||||
enum: uniqueEnums,
|
||||
};
|
||||
if (schema.description) {
|
||||
result.description = schema.description;
|
||||
}
|
||||
return result;
|
||||
}
|
||||
}
|
||||
|
||||
const sanitized: any = {};
|
||||
|
||||
for (const [key, value] of Object.entries(schema)) {
|
||||
// skip $schema field entirely
|
||||
if (key === "$schema") {
|
||||
continue;
|
||||
}
|
||||
|
||||
// skip any_of if we already converted it above
|
||||
if (key === "anyOf" && schema.anyOf) {
|
||||
continue;
|
||||
}
|
||||
|
||||
// convert $defs to definitions for draft-07 compatibility
|
||||
if (key === "$defs") {
|
||||
sanitized.definitions = sanitizeSchema(value);
|
||||
continue;
|
||||
}
|
||||
|
||||
// recursively sanitize nested objects
|
||||
sanitized[key] = sanitizeSchema(value);
|
||||
}
|
||||
|
||||
return sanitized;
|
||||
}
|
||||
|
||||
/**
|
||||
* Wrap a schema to sanitize its JSON Schema output for Gemini/OpenCode compatibility.
|
||||
* xsschema calls ~standard.jsonSchema.input() for schemas that implement StandardJSONSchemaV1
|
||||
* (i.e. have ~standard.jsonSchema), which includes arktype and our AJV-backed JSON schema wrapper.
|
||||
* Schemas without ~standard.jsonSchema are returned unchanged (sanitization skipped).
|
||||
*/
|
||||
function wrapSchema(
|
||||
schema: StandardSchemaV1<any> & {
|
||||
"~standard": Partial<StandardJSONSchemaV1<any>["~standard"]>;
|
||||
}
|
||||
): StandardSchemaV1<any> {
|
||||
const standardProps = schema["~standard"];
|
||||
|
||||
if (!("jsonSchema" in standardProps)) {
|
||||
return schema;
|
||||
}
|
||||
|
||||
const jsonSchema = standardProps.jsonSchema;
|
||||
const wrapped: StandardSchemaV1<any> & StandardJSONSchemaV1<any> = {
|
||||
...schema,
|
||||
"~standard": {
|
||||
...standardProps,
|
||||
jsonSchema: {
|
||||
input: (options) => sanitizeSchema(jsonSchema.input(options)),
|
||||
output: (options) => sanitizeSchema(jsonSchema.output(options)),
|
||||
},
|
||||
},
|
||||
};
|
||||
return wrapped;
|
||||
}
|
||||
|
||||
/**
|
||||
* Transform tool to sanitize its parameter schema for Gemini CLI compatibility
|
||||
*/
|
||||
function sanitizeTool<T extends Tool<any, any>>(tool: T): T {
|
||||
if (!tool.parameters) {
|
||||
return tool;
|
||||
}
|
||||
|
||||
const wrappedSchema = wrapSchema(tool.parameters);
|
||||
|
||||
// create a new tool with wrapped schema
|
||||
return {
|
||||
...tool,
|
||||
parameters: wrappedSchema,
|
||||
} as T;
|
||||
}
|
||||
|
||||
export const addTools = (ctx: ToolContext, server: FastMCP<any>, tools: Tool<any, any>[]) => {
|
||||
// sanitize schemas for gemini agent and opencode (when using Google API)
|
||||
// both have issues with draft-2020-12 schemas and any_of enum constructs
|
||||
const shouldSanitize = ctx.agent.name === "gemini" || ctx.agent.name === "opencode";
|
||||
|
||||
export const addTools = (_ctx: ToolContext, server: FastMCP<any>, tools: Tool<any, any>[]) => {
|
||||
for (const tool of tools) {
|
||||
const processedTool = shouldSanitize ? sanitizeTool(tool) : tool;
|
||||
server.addTool(processedTool);
|
||||
server.addTool(tool);
|
||||
}
|
||||
return server;
|
||||
};
|
||||
|
||||
Reference in New Issue
Block a user