[Dong1017]

Governing Issue and delivery slice

Refs #372
Refs #362

• Accepted scope / delegation: After the docs: record the target module layout and import gate #376 directory rule and the test: add a fake prepare-to-assess handoff loop #377/test: connect one CPU offline read, record, and assessment #378 connection tests, this is the conservative fixture-path cleanup from the suggested PR4 row. It moves two fixture trees and updates in-repo citations. It does not treat the 3+1 split as a newly accepted plan.
• What this PR delivers: docs/examples/skill-comparison/ now lives at tests/fixtures/skill-comparison/, and docs/examples/skill-eval-pr/ now lives at tests/fixtures/skill-eval-pr/. Every in-repo citation of the old paths is updated in the same change.
• What it does not deliver: No production Apps package, no knowledge-content change, and no campaign acceptance. docs/examples/harness-improvement/ stays. docs/plans/, PHASE1_REPORT.md, SUBMISSION_NOTES.md, evidence/, docs/reviews/, and skill-library content stay. Existing helpers skill_eval_pr_support.py, skill_eval_compare_support.py, skill_eval_pr_demo.py, skill_eval_compare_demo.py, and tests/fixtures/skill_eval_engineering/ are unchanged. cli.py, arguments.py, harness/runner.py, and pyproject.toml are untouched. Issues 372 and 362 stay open. This is not Issue acceptance.

Knowledge authorship declaration

Not applicable. This PR does not change protected knowledge paths, knowledge content, or Bundle revision rules. No policy exception is requested.

{
  "schema_version": 1,
  "head_sha": "bac4f895063cf7dec827dd0af8d4c14f78fd9b31",
  "human_only": false,
  "material_ai_authors": [
    {
      "tool": "Cursor",
      "model_id": "grok-4.7",
      "effort": "unknown",
      "role": "author"
    }
  ]
}

Changed artifacts and evidence

Moved with git mv. File bytes are unchanged except the one relative link called out below.

Old path	New path
docs/examples/skill-comparison/addition.json	tests/fixtures/skill-comparison/addition.json
docs/examples/skill-comparison/cases.json	tests/fixtures/skill-comparison/cases.json
docs/examples/skill-comparison/pr133-diagnostic-case05.json	tests/fixtures/skill-comparison/pr133-diagnostic-case05.json
docs/examples/skill-comparison/report-example.json	tests/fixtures/skill-comparison/report-example.json
docs/examples/skill-comparison/report-example.md	tests/fixtures/skill-comparison/report-example.md
docs/examples/skill-comparison/update.json	tests/fixtures/skill-comparison/update.json
docs/examples/skill-eval-pr/summary-example.md	tests/fixtures/skill-eval-pr/summary-example.md

Updated references:

Location	What changed
tests/test_skill_eval_feedback.py	TASK_PATH now reads tests/fixtures/skill-comparison/pr133-diagnostic-case05.json
docs/skill-eval-comparison.md	Links to addition.json, update.json, report-example.md, and report-example.json, plus the prepare --config path
docs/skill-eval-pr.md	Link to summary-example.md
.github/workflows/skill-eval-pr.yml	pull_request_target path filter now watches tests/fixtures/skill-comparison/**
tools/skill_eval_pr_cases.json	cases for inference-benchmark-design, audit-fixed-workload-claim, and optimization-search-policy
docs/module-layout.md	"Later fixture move" is now "Fixture locations" and names the current directories
tests/fixtures/skill-comparison/report-example.md	Relative link to docs/skill-eval-comparison.md updated for the new depth

tests/fixtures/skill-eval-pr/summary-example.md still links to tests/test_skill_eval_pr.py and tests/fixtures/skill_eval_pr_demo.py. Those relative links were left as they were because the directory depth is unchanged and both targets still resolve. docs/examples/harness-improvement/summary.json remains the only file under docs/examples/.

git grep on the committed tree for docs/examples/skill-comparison, docs/examples/skill-eval-pr, examples/skill-comparison, and examples/skill-eval-pr returned no hits. Past Issue and PR comments were not edited.

Requirement	Actual change	Check / evidence and result	Remaining gap
Move both fixture trees	Seven git mv renames listed above	Rename similarity 100% except report-example.md at 98% for the relative link	None in this diff
Update in-repo citations	Seven reference edits listed above	git grep for the old path strings: no hits	External Issue/PR comments can still name the old paths
Tests still read the fixtures	Feedback test path and mapping cases paths	21 passed in 0.70s (command below)	Full pytest was not run
Workflow path filter still loads	.github/workflows/skill-eval-pr.yml	Included in the 21 tests via the two workflow YAML checks	No new path filter was added for tests/fixtures/skill-eval-pr/**; the old workflow did not watch that tree
Style of the touched Python file	tests/test_skill_eval_feedback.py	ruff check passed (ruff 0.17.4)	Full-repo ruff was not run

Checks and dependencies

• Checks actually run (result and tested scope): On Git 2.56.0 (the snapshot reader requires --no-lazy-fetch, Git >= 2.45): python3 -m pytest tests/test_skill_eval_feedback.py tests/test_skill_eval_pr.py::test_maintained_mapping_selects_real_acceptance_paths tests/test_skill_eval_pr.py::test_workflow_separates_execution_and_publication_credentials tests/test_skill_eval_auto.py::test_workflow_is_opt_in_and_controller_is_only_budget_writer tests/test_skill_eval_planning_boundaries.py::test_expired_paid_configuration_preserves_provider_plan_and_free_publication --override-ini addopts= -q --tb=no — 21 passed in 0.70s. That set reads the moved diagnostic case, selects real acceptance cases through tools/skill_eval_pr_cases.json, and loads the edited workflow YAML. python3 -m ruff check tests/test_skill_eval_feedback.py — All checks passed (ruff 0.17.4). git diff --check HEAD~1 — clean. git grep for the four old path strings — no hits.
• Failed, skipped, timed-out or not-run checks (if applicable): The pytest run had no failed, skipped, or timed-out tests. Full pytest was not run. ruff check src tests and ruff check src tests tools were not run; the only touched Python file was checked. Ready CI, hardware, GPU/NPU, and model runs were not run. A green check is not campaign or effect acceptance.
• Dependencies / shared interfaces / integration assumptions (if applicable): Open feat: support shared YAML task input #367 still owns cli.py, arguments.py, harness/runner.py, and pyproject.toml. This branch does not edit them. task: Apps 应用层详细设计与迁移计划（CLI / Python / Plugin） #371 still names the Apps package. No merge gate was added to feat: support shared YAML task input #367 or feat: share RSI optimization management across project and benchmark tasks #338.

Remaining work and Issue closure

• Remaining Issue requirements / follow-up: The suggested PR4 row also mentioned an archive review of docs/plans/, PHASE1_REPORT.md, and SUBMISSION_NOTES.md. Those files were not moved. Production Apps assembly, S2/S3/S4 production work, and the undecided contract items stay outside this diff.
• Goal-acceptance owner or accepted closure delegation: Not claimed.
• Goal ready for acceptance? Not claimed

Handoff (only when an actual handoff or decision changed)

• Accepted decision / status and link: discussion: Knowledge / Harness / Benchmark 三模块目标架构——现状、原语、交接时序与遗留问题 #362 A–D remains the architecture decision. docs: record the target module layout and import gate #376 recorded the target fixture directory names. This PR applies that path cleanup only.
• Recipient and communication status: Review requested from zengchuang-hw and amy-why-3459 on head bac4f895063cf7dec827dd0af8d4c14f78fd9b31.
• Application status and artifact/evidence: pending review
• Relevant workspace state and next action: Do not merge from this change set. Issues 372 and 362 stay open.

  

[cursor]

@zengchuang-hw @amy-why-3459 Please review this fixture-path cleanup. Head SHA: bac4f895063cf7dec827dd0af8d4c14f78fd9b31. Issues 372 and 362 stay open. This is not Issue acceptance.

[github-actions]

Skill evaluation

PLANNED: not executed; there are no task outcomes yet.
No approval, model efficacy or current-head test result is implied.

• Next: resolve check applicability at head bac4f895063cf7dec827dd0af8d4c14f78fd9b31.
• Task selection: no supported automatic comparison; choose a relevant source/content check and disposition uncovered claims.
• Selected comparisons: 0 · Estimated task calls: 0. Uncovered/unsupported changes: 20; Canonical library paths: 0; Generated projection paths: 0; Other paths: 20. These are paths, not independent claims.
• Return: command, input/source pin, exit status, outputs/failures and untested claims for independent verified/refuted/unverified disposition.

Author verification steps (nonpaid; not yet run)

1. Bind the checkout. Run git rev-parse HEAD; it must equal bac4f895063cf7dec827dd0af8d4c14f78fd9b31. If it differs, obtain a current plan; do not reuse old results.
2. Choose an applicable check. Match each changed behavior to the task below or a source/content check; a path match alone does not establish relevance.
   No supported automatic comparison is selected. Review the canonical changes and explain which claims need evidence, are unaffected, or remain unverified; do not treat generated records as separate behavioral claims.
   Uncovered scope by location: Canonical library paths: 0; Generated projection paths: 0; Other paths: 20. Path counts are not claim counts; no metadata-only conclusion is inferred.
3. Return evidence for independent review: head, check/task/source pin, command, environment, exit status, outputs/failures and untested claims. Label scripted controls. Request each claim's disposition: verified, refuted or unverified; this plan proves neither model provenance nor benefit.
   Check recipes and evidence limits

Evidence and logs · Retention: 30 days.
Retained artifacts are verifiable only before Unix time 1796874745 while the linked artifact is available. Expired evidence is unverified.

Plan bindings, budget, frozen exam and advisory details

Advisory: obtain missing evidence

Retained evidence availability: planned / unexecuted. No approval or merge authorization is issued.
Applicable only while these revisions and inputs match the current PR.
Paid execution requires trusted label authorization and cumulative PR reservation. Unknown accounting blocks new paid calls; planning is not execution.
Pipeline evidence only; offline fixed outputs do not establish skill efficacy.
Unchanged-correct is valid neutral evidence, not incremental benefit.
Content/source review and authorship admission remain separate.
Trusted regrading checks retained answers, not model/hardware provenance or a hostile executor; this is not an efficacy certification.

• Retained evidence is planned / unexecuted.
• No supported comparison cases were selected.
• Uncovered/unsupported scope needs a case or explicit reviewer disposition.
  Plan SHA-256: 8189f18d40203770b80391ab9520ed72b0599c210b46dfed1b9a0e5dac55e6da
  Base: b8df3715709b806d06c4bfc9158ab41b18ded11e · Head: bac4f895063cf7dec827dd0af8d4c14f78fd9b31
  Trusted implementation: b8df3715709b806d06c4bfc9158ab41b18ded11e

Model/settings: {"effort":"not_applicable","name":"claude-opus-5-5"}
Budget: {"comparisons":0,"estimated_calls":0,"limits_per_comparison":{"campaign_seconds":600,"max_task_calls":2,"repeats":1,"timeout_seconds":180,"token_stop_new_trials":200000},"provider_config_sha256":"b28d80f8194fbe0f5a9f01515efb24b218979fd45f702776560da276f4b30d04","reservation":null}

Skill	Comparison	Cases	Estimated calls

Automatic execution / cumulative PR budget:

{"reason":"Trusted configuration or pricing missing, invalid or expired","remaining_usd":"unknown","reserved_or_unresolved_usd":"unknown","spent_or_legacy_usd":"unknown","state":"blocked"}

Provider/settings: {"client":{"api_version":"2023-06-01","endpoint":"https://yibuapi.com/v1/messages","kind":"anthropic-messages-full-context-v1","max_output_tokens":4096,"max_read_bytes":32768,"max_request_bytes":262144,"max_requests":1,"max_tool_calls":12,"provider":"yibuapi"},"limits":{"token_stop_new_trials":200000},"model":{"effort":"not_applicable","name":"claude-opus-5-5"},"pricing":{"currency":"USD","expires_at":1793271228,"group":"default","max_input_tokens":1000000,"model":"claude-opus-5-5","rate_evidence":"https://github.com/InferMatrixLab/InferRSI/issues/215#issuecomment-6467531341","rates":{"cache_read":"0.4","cache_write":"10","input":"8","output":"40"},"request_fee":"0","unit":"per_million_tokens"},"spend_limit":{"amount":"200","currency":"USD"}}

Uncovered / unsupported paths (20)

• uncovered: ".github/workflows/skill-eval-pr.yml"
• uncovered: "docs/examples/skill-comparison/addition.json"
• uncovered: "docs/examples/skill-comparison/cases.json"
• uncovered: "docs/examples/skill-comparison/pr133-diagnostic-case05.json"
• uncovered: "docs/examples/skill-comparison/report-example.json"
• uncovered: "docs/examples/skill-comparison/report-example.md"
• uncovered: "docs/examples/skill-comparison/update.json"
• uncovered: "docs/examples/skill-eval-pr/summary-example.md"
• uncovered: "docs/module-layout.md"
• uncovered: "docs/skill-eval-comparison.md"
• uncovered: "docs/skill-eval-pr.md"
• uncovered: "tests/fixtures/skill-comparison/addition.json"
• uncovered: "tests/fixtures/skill-comparison/cases.json"
• uncovered: "tests/fixtures/skill-comparison/pr133-diagnostic-case05.json"
• uncovered: "tests/fixtures/skill-comparison/report-example.json"
• uncovered: "tests/fixtures/skill-comparison/report-example.md"
• uncovered: "tests/fixtures/skill-comparison/update.json"
• uncovered: "tests/fixtures/skill-eval-pr/summary-example.md"
• uncovered: "tests/test_skill_eval_feedback.py"
• uncovered: "tools/skill_eval_pr_cases.json"

[cursor]

LGTM

PR review verdict — PASS
Reviewed base b8df371 and head bac4f89 (full diff, opened, no prior review).

Verified:

• Renames move docs/examples/skill-comparison/ to tests/fixtures/skill-comparison/ (addition.json, update.json, cases.json, pr133-diagnostic-case05.json, report-example.json, report-example.md) and docs/examples/skill-eval-pr/summary-example.md to tests/fixtures/skill-eval-pr/summary-example.md. The JSON fixtures and the summary excerpt are byte-identical renames.
• tests/fixtures/skill-comparison/report-example.md backlink is ../../../docs/skill-eval-comparison.md, which resolves to docs/skill-eval-comparison.md. The sibling report-example.json link stays in the same directory.
• docs/skill-eval-comparison.md markdown links and the skill_eval_compare prepare --config example use tests/fixtures/skill-comparison/. Links in docs/*.md use ../tests/fixtures/... from docs/.
• docs/skill-eval-pr.md points the offline summary excerpt at ../tests/fixtures/skill-eval-pr/summary-example.md.
• tests/test_skill_eval_feedback.py loads pr133-diagnostic-case05.json with Path(__file__).parents[1] (repository root) under tests/fixtures/skill-comparison/.
• tools/skill_eval_pr_cases.json keeps the ../ prefix used from tools/, so ../tests/fixtures/skill-comparison/... resolves to tests/fixtures/skill-comparison/....
• .github/workflows/skill-eval-pr.yml path filter lists tests/fixtures/skill-comparison/** where it previously listed docs/examples/skill-comparison/**.
• docs/module-layout.md records both fixture directories and still says not to overwrite skill_eval_pr_support.py and skill_eval_compare_support.py.

Not verified here: tests were not executed and CI was not checked. Rename-only fixture bodies had no hunks, so embedded path strings inside those files were not re-read beyond this patch.

[amy-why-3459]

LGTM

PR review verdict — PASS
Reviewed base b8df371 and head bac4f89 (new PR, full diff).

Verified:

• Seven renames from docs/examples/skill-comparison/ and docs/examples/skill-eval-pr/ to tests/fixtures/skill-comparison/ and tests/fixtures/skill-eval-pr/; only report-example.md changes content (backlink now ../../../docs/skill-eval-comparison.md, resolves correctly).
• Reference updates in tests/test_skill_eval_feedback.py (TASK_PATH), tools/skill_eval_pr_cases.json (../tests/fixtures/... relative to tools/), .github/workflows/skill-eval-pr.yml path filter, docs/skill-eval-comparison.md, docs/skill-eval-pr.md, and docs/module-layout.md.
• Code search on main for skill-comparison, docs/examples and summary-example only hits files this PR updates, so no stale reference to the old paths is left behind.
• CI on this head is green: lint-and-smoke, full-check (3.10/3.11), installed-package (3.10/3.11), pr-references, declarations, skill-eval plan/publish.

Note (non-blocking): the workflow uses pull_request_target, so the new path filter only takes effect after merge.