ci: enforce Astryx inventory for code changes

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.

[liuxiaocs7]

Summary

While validating the Desktop build fix in #3643, we found that the generated Astryx surface inventory still reported 204 files even though the current path list and table contain 206. Because the gate was selected only for direct Astryx-surface changes, repository-wide drift could escape ordinary code validation.

• Refresh the generated total to 206 files.
• Run the Astryx inventory for every code validation while preserving the inventory-only path.
• Place the inexpensive gate before Build so drift is reported directly.
• Add workflow contract regression coverage.

Fixes#3646

Verification

Passed locally:

• node --test scripts/ci-test-plan.test.mjs (30/30)
• npm run astryx:surface-inventory (206 files)
• npm run lint
• npm run format:check
• npm run build
• npm run typecheck
• git diff --check

Also attempted NODE_NO_WARNINGS=1 npm test. The suites covering this change passed, while the repository-wide run remained non-green on three unrelated checks: an existing UI deferred-activation assertion, a Runtime Host real-model terminal timeout, and a Python test that uses Python 3.10 union syntax on the local Python 3.9.6 runtime.

AI use

Select exactly one:

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: OpenAI Codex helped diagnose the CI coverage gap, implement the workflow and regression-test changes, and draft the issue and PR text.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[liuxiaocs7]

Hi, @Astro-Han, could you help take a look when free, thanks!

[Astro-Han]

Reviewed at exact head f046172ab9661d648116394c6887ae4368eb0bd9. No P0–P3 findings. The gate remains scoped by the existing planner: ordinary documentation-only changes keep code=false and do not run it, inventory artifacts run through astryx_surface, and code changes run the repository-wide regenerate-and-diff check before Build. The committed 206-file inventory regenerates byte-for-byte; planner tests pass 30/30, and the exact-head test check is terminal green.