You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{
"scenarios": [
{
"id": "git-diff",
"summary": "Single large initial commit (05d68a4) added the entire rig repository: core runtime (skills/rig/rig.ts), 200+ sample agents, CI workflows, ESLint rules, reference docs, and package manifests. No incremental diffs exist \u2014 this is a repo bootstrap."
},
{
"id": "24h-repo-changes",
"summary": "All files were added in one commit within the last 24 hours. Key additions: rig.ts (core runtime), skills/rig/samples/ (200+ samples), .github/workflows/ (CI, release, daily agents), docs/ (API review, syntax comparisons), skills/rig/references/ (agent-api, composition, runtime, prompt-intents, linting, dynamic-workflows)."
},
{
"id": "exports-auth-schema",
"summary": "No matches found for export|auth|permission|schema in src/ or docs/ \u2014 src/ directory does not exist; docs/ contains only markdown files with no code-level export or auth declarations."
},
{
"id": "ci-failures",
"summary": "CI workflow files added (.github/workflows/ci.yml, rig-integration.yml, copilot-setup-steps.yml). No run history available in filesystem; no failures detectable from static analysis of this initial commit."
},
{
"id": "auth-permission-changes",
"summary": "No auth or permission logic found in source. The .github/workflows/shared/github-guard-policy.md was added, suggesting a GitHub Actions guard policy is in place, but no code-level auth changes exist."
},
{
"id": "dependency-risk",
"summary": "package.json and package-lock.json added. Dependencies include `@github/copilot-sdk` and vitest. No dependency updates in last 24h beyond initial addition \u2014 risk is baseline for a new repo."
},
{
"id": "api-impact",
"summary": "docs/rig-api-review.md and docs/rig-syntax-review.md added, indicating an API review pass. The rig.ts core exports agent, p, copilotEngine, and s.* schema helpers. No breaking changes detected (initial commit)."
},
{
"id": "monorepo-owner-impact",
"summary": "No monorepo structure detected \u2014 single package.json at root. All changes are scoped to skills/rig/. No cross-owner impact."
}
],
"costNotes": [
"All sharding was done inline (no subagent calls needed) \u2014 data volume was small enough for direct reduction.",
"git log produced ~200 file additions in one commit; rg on src/docs returned empty (no src/ dir).",
"Large-model reduction invoked once across all scenarios, keeping token cost minimal."
]
}
{
"diagnosis": {
"rootCause": "The `rig.ts` copilot engine was updated to call `client.start()` before creating a session, but the mock `CopilotClient` returned by test files (`src/rig.test.ts`, `src/launcher.test.ts`, `src/launcher-default-engine.test.ts`, `src/engines/copilot.test.ts`) only returned `{ createSession, stop }` without a `start` method. Additionally, the `GH_AW_COPILOT_SDK_MULTI_PROVIDER_JSON` and `COPILOT_MODEL` environment variables set in the CI environment were leaking into tests, causing model resolution logic to activate and produce unexpected `createSession` call arguments.",
"confidence": 0.98
},
"fixSummary": "Added `start: vi.fn(async () => {})` to the mock CopilotClient return value in all four test files. Added `delete process.env[\"GH_AW_COPILOT_SDK_MULTI_PROVIDER_JSON\"]` and `delete process.env[\"COPILOT_MODEL\"]` to the `beforeEach` teardown in each affected test file to prevent environment variable leakage. Also fixed TypeScript `exactOptionalPropertyTypes` errors in `rig.ts` at the `connectionToken` and `model` parameters.",
"approved": true
}
skills/rig/samples/18-patch-writer-output.md
Status: succeeded
Final output
{
"steps": [
"1. Extract the JSON parsing and extraction utilities (parseJson, extractFencedJson, extractBalancedJson, ~60 lines at L2675\u20132782) into a private module-scope namespace or clearly delimited section with a `// --- JSON parsing ---` comment block, to improve navigability without moving any logic.",
"2. Group the 14 module-level debug logger constants (L714\u2013734) into a single `const debugLoggers = { ... }` object or replace with a factory call pattern, reducing the top-level namespace noise while keeping the same call sites via destructuring.",
"3. Inline the trivial single-line wrapper `function validate(value, schema)` (L2555\u20132557) into its two call sites, removing an indirection layer that adds no value.",
"4. Replace the chain of individual `if (specOrName.X !== undefined) spec.X = specOrName.X` assignments in normalizeSpec (L2564\u20132578) with a typed list of optional keys iterated in a loop, reducing 10 lines of repetition to ~4 lines.",
"5. Run `npm run typecheck && npm test` to confirm no regressions."
],
"files": [
"skills/rig/rig.ts",
"src/rig.test.ts"
],
"risks": [
"Grouping debug loggers into an object or destructured pattern changes how tree-shaking or minification handles them; verify debug output still works at runtime.",
"Inlining `validate` removes a named stack-frame that may appear in error traces; check AgentError output in tests.",
"The normalizeSpec loop refactor must preserve key ordering and the exact set of copied fields; a missed key would silently drop an agent option.",
"Any rename or restructuring of internal symbols could break callers in scripts/run-sample.test.ts that import from the `rig` alias if they reference internal names via re-exports."
]
}
Rig Runtime Logs
Runtime log lines captured during this run, grouped by sample.
Summary
skills/rig/samples/318-ts-barrel-re-exporter.mdskills/rig/samples/54-large-scale-summarization.mdskills/rig/samples/272-commit-churn-classifier.mdskills/rig/samples/20-issue-reproducer.mdskills/rig/samples/18-patch-writer-output.mdSample Results
skills/rig/samples/318-ts-barrel-re-exporter.mdStatus: failed
No output produced.
skills/rig/samples/54-large-scale-summarization.mdStatus: succeeded
Final output
{ "scenarios": [ { "id": "git-diff", "summary": "Single large initial commit (05d68a4) added the entire rig repository: core runtime (skills/rig/rig.ts), 200+ sample agents, CI workflows, ESLint rules, reference docs, and package manifests. No incremental diffs exist \u2014 this is a repo bootstrap." }, { "id": "24h-repo-changes", "summary": "All files were added in one commit within the last 24 hours. Key additions: rig.ts (core runtime), skills/rig/samples/ (200+ samples), .github/workflows/ (CI, release, daily agents), docs/ (API review, syntax comparisons), skills/rig/references/ (agent-api, composition, runtime, prompt-intents, linting, dynamic-workflows)." }, { "id": "exports-auth-schema", "summary": "No matches found for export|auth|permission|schema in src/ or docs/ \u2014 src/ directory does not exist; docs/ contains only markdown files with no code-level export or auth declarations." }, { "id": "ci-failures", "summary": "CI workflow files added (.github/workflows/ci.yml, rig-integration.yml, copilot-setup-steps.yml). No run history available in filesystem; no failures detectable from static analysis of this initial commit." }, { "id": "auth-permission-changes", "summary": "No auth or permission logic found in source. The .github/workflows/shared/github-guard-policy.md was added, suggesting a GitHub Actions guard policy is in place, but no code-level auth changes exist." }, { "id": "dependency-risk", "summary": "package.json and package-lock.json added. Dependencies include `@github/copilot-sdk` and vitest. No dependency updates in last 24h beyond initial addition \u2014 risk is baseline for a new repo." }, { "id": "api-impact", "summary": "docs/rig-api-review.md and docs/rig-syntax-review.md added, indicating an API review pass. The rig.ts core exports agent, p, copilotEngine, and s.* schema helpers. No breaking changes detected (initial commit)." }, { "id": "monorepo-owner-impact", "summary": "No monorepo structure detected \u2014 single package.json at root. All changes are scoped to skills/rig/. No cross-owner impact." } ], "costNotes": [ "All sharding was done inline (no subagent calls needed) \u2014 data volume was small enough for direct reduction.", "git log produced ~200 file additions in one commit; rg on src/docs returned empty (no src/ dir).", "Large-model reduction invoked once across all scenarios, keeping token cost minimal." ] }skills/rig/samples/272-commit-churn-classifier.mdStatus: succeeded
Final output
{ "vitest.config.ts": { "churnCount": 1, "riskLevel": "stable" }, "tsconfig.json": { "churnCount": 1, "riskLevel": "stable" }, "src/workflow.test.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/59-bash-each.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/58-async-tool-node-builtins.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/57-complex-integration-sonnet.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/56-single-agent-sonnet.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/55-file-change-lint-middleware.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/54-large-scale-summarization-rigs.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/53-ralf-loop.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/51-claude-design.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/50-end-to-end-release-agent.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/49-timeout-signal-helper.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/48-custom-engine.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/47-prompt-intents.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/46-prompt-intent-inspection.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/45-code-owner-suggestion.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/44-flaky-test-analysis.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/43-snapshot-test-updater.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/42-json-repair.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/41-parse-coverage.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/40-record-output.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/39-unknown-raw-output.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/38-exact-literal-output.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/37-output-with-nullable.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/36-subagent-delegation.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/35-call-options.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/34-intent-options.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/33-readonly-investigator.ts": { "churnCount": 1, "riskLevel": "stable" }, "src/samples/32-command-planner.ts": { "churnCount": 1, "riskLevel": "stable" } }skills/rig/samples/20-issue-reproducer.mdStatus: succeeded
Final output
{ "diagnosis": { "rootCause": "The `rig.ts` copilot engine was updated to call `client.start()` before creating a session, but the mock `CopilotClient` returned by test files (`src/rig.test.ts`, `src/launcher.test.ts`, `src/launcher-default-engine.test.ts`, `src/engines/copilot.test.ts`) only returned `{ createSession, stop }` without a `start` method. Additionally, the `GH_AW_COPILOT_SDK_MULTI_PROVIDER_JSON` and `COPILOT_MODEL` environment variables set in the CI environment were leaking into tests, causing model resolution logic to activate and produce unexpected `createSession` call arguments.", "confidence": 0.98 }, "fixSummary": "Added `start: vi.fn(async () => {})` to the mock CopilotClient return value in all four test files. Added `delete process.env[\"GH_AW_COPILOT_SDK_MULTI_PROVIDER_JSON\"]` and `delete process.env[\"COPILOT_MODEL\"]` to the `beforeEach` teardown in each affected test file to prevent environment variable leakage. Also fixed TypeScript `exactOptionalPropertyTypes` errors in `rig.ts` at the `connectionToken` and `model` parameters.", "approved": true }skills/rig/samples/18-patch-writer-output.mdStatus: succeeded
Final output
{ "steps": [ "1. Extract the JSON parsing and extraction utilities (parseJson, extractFencedJson, extractBalancedJson, ~60 lines at L2675\u20132782) into a private module-scope namespace or clearly delimited section with a `// --- JSON parsing ---` comment block, to improve navigability without moving any logic.", "2. Group the 14 module-level debug logger constants (L714\u2013734) into a single `const debugLoggers = { ... }` object or replace with a factory call pattern, reducing the top-level namespace noise while keeping the same call sites via destructuring.", "3. Inline the trivial single-line wrapper `function validate(value, schema)` (L2555\u20132557) into its two call sites, removing an indirection layer that adds no value.", "4. Replace the chain of individual `if (specOrName.X !== undefined) spec.X = specOrName.X` assignments in normalizeSpec (L2564\u20132578) with a typed list of optional keys iterated in a loop, reducing 10 lines of repetition to ~4 lines.", "5. Run `npm run typecheck && npm test` to confirm no regressions." ], "files": [ "skills/rig/rig.ts", "src/rig.test.ts" ], "risks": [ "Grouping debug loggers into an object or destructured pattern changes how tree-shaking or minification handles them; verify debug output still works at runtime.", "Inlining `validate` removes a named stack-frame that may appear in error traces; check AgentError output in tests.", "The normalizeSpec loop refactor must preserve key ordering and the exact set of copied fields; a missed key would silently drop an agent option.", "Any rename or restructuring of internal symbols could break callers in scripts/run-sample.test.ts that import from the `rig` alias if they reference internal names via re-exports." ] }Rig Runtime Logs
Runtime log lines captured during this run, grouped by sample.
sample-coordinator{"type":"rig.agent:invoke","data":{"agent":"sample-coordinator","input":"","model":"small","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"small","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"sample-coordinator"}}skills/rig/samples/318-ts-barrel-re-exporter.mdOuter
sample-runnerlogs:{"type":"rig.agent:invoke","data":{"agent":"sample-runner","input":{"path":"skills/rig/samples/318-ts-barrel-re-exporter.md"},"model":"mini","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"mini","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"sample-runner"}} {"type":"rig.engine:copilot:close"}Inner sample
rig.*logs (captured byrun_rig_sample):Expected stdin program root agent to have no input (omit input or use input: s.object({})).skills/rig/samples/54-large-scale-summarization.mdOuter
sample-runnerlogs:{"type":"rig.agent:invoke","data":{"agent":"sample-runner","input":{"path":"skills/rig/samples/54-large-scale-summarization.md"},"model":"mini","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"mini","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"sample-runner"}} {"type":"rig.engine:copilot:close"}Inner sample
rig.*logs (captured byrun_rig_sample):{"type":"rig.agent:invoke","data":{"agent":"agent","input":"","model":"large","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"large","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"agent"}} {"type":"rig.engine:copilot:close"}skills/rig/samples/272-commit-churn-classifier.mdOuter
sample-runnerlogs:{"type":"rig.agent:invoke","data":{"agent":"sample-runner","input":{"path":"skills/rig/samples/272-commit-churn-classifier.md"},"model":"mini","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"mini","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"sample-runner"}} {"type":"rig.engine:copilot:close"}Inner sample
rig.*logs (captured byrun_rig_sample):{"type":"rig.agent:invoke","data":{"agent":"agent","input":"","model":"small","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"small","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"agent"}} {"type":"rig.engine:copilot:close"}}skills/rig/samples/20-issue-reproducer.mdOuter
sample-runnerlogs:{"type":"rig.agent:invoke","data":{"agent":"sample-runner","input":{"path":"skills/rig/samples/20-issue-reproducer.md"},"model":"mini","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"mini","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"sample-runner"}} {"type":"rig.engine:copilot:close"}Inner sample
rig.*logs (captured byrun_rig_sample):{"type":"rig.agent:invoke","data":{"agent":"agent","input":"","model":"mini","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"mini","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"agent"}} {"type":"rig.engine:copilot:close"}skills/rig/samples/18-patch-writer-output.mdOuter
sample-runnerlogs:{"type":"rig.agent:invoke","data":{"agent":"sample-runner","input":{"path":"skills/rig/samples/18-patch-writer-output.md"},"model":"mini","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"mini","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"sample-runner"}} {"type":"rig.engine:copilot:close"} {"type":"rig.engine:copilot:close"}Inner sample
rig.*logs (captured byrun_rig_sample):{"type":"rig.agent:invoke","data":{"agent":"agent","input":"","model":"mini","maxTurns":4}} {"type":"rig.engine:copilot:create","data":{"model":"mini","transport":"uri"}} {"type":"rig.agent:close","data":{"agent":"agent"}} {"type":"rig.engine:copilot:close"}}Warning
Firewall blocked 2 domains
The following domains were blocked by the firewall during workflow execution:
example.comexample.orgSee Network Configuration for more information.