[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7E8DSJYVHSN5TGCM9JEVT2P":3,"skill-same:01M7E8DSJYVHSN5TGCM9JEVT2P":61},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":33,"commitSha":34,"skillDirectory":35,"forksCount":14,"repositoryHtmlUrl":36,"repositoryHomepageUrl":6,"ownerHtmlUrl":37,"ownerBlogUrl":38,"ownerDescription":39,"updatedAt":40,"install":41,"resources":54},"01M7E8DSJYVHSN5TGCM9JEVT2P","pytest-optimizer",null,"Use when pytest tests or fixtures are slow and you want them faster: speed up the suite with measured, safety-gated optimization.","tony\u002Fskills","tony","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F26336?v=4","01M72E5RNJ7SQV8F9KBM5RKRMN","Tony Narlock","01M7E7MBN5C6YZ4JJJ5Y82KBJG",0,"MIT","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fpytest-optimizer\u002Fskills\u002Fpytest-optimizer\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: pytest-optimizer\ndescription: \"Use when pytest tests or fixtures are slow and you want them faster: speed up the suite with measured, safety-gated optimization.\"\nallowed-tools: [\"Bash\", \"Read\", \"Grep\", \"Glob\", \"Write\", \"Edit\", \"AskUserQuestion\"]\nargument-hint: \"[test-path-or-marker] [--phase=scan|benchmark|plan|execute] [--force] [--hygiene] [--memory-dir=\u003Cpath>]\"\n---\n\n# pytest-optimizer\n\nProfile a pytest suite, find safe speedups, and apply each as its own verified\ncommit on a new branch, through a resumable, idempotent four-phase loop.\n\nMeasure before you mutate, and prove a speedup against the project's own timing\nnoise before you trust it. A change whose measured saving falls inside the noise\nband, whose run exits non-zero, or whose collected tests change is never\ncommitted, except a deselection the user approved with its test count, which is\nreported as a smaller default run, not a faster suite.\n\n`$ARGUMENTS` may name a test path or marker to scope the run, a `--phase` to jump\nto, `--force` to ignore idempotency tokens and re-run, `--hygiene` to add the opt-in typing\nand parametrize detectors, and `--memory-dir` to\noverride where durable state is written.\n\nPaths under `..\u002F..\u002Fscripts\u002F` resolve from this skill's directory: print the absolute path once and write it literally in each command, since shell variables do not persist between Bash calls.\n\n## The loop\n\n```\n00-scan  ->  01-benchmark  ->  02-plan  ->  03-execute  ->  (re-scan)\n```\n\nEach phase reads and writes durable JSON under the resolved memory directory;\nre-running resumes at the first phase whose `completed` flag is false, which\n`measure.py state resume` reports. After `03-execute`, re-scan and loop until no\ncandidate clears the rubric.\n\nThe phase skills set `disable-model-invocation`, so this skill cannot invoke\nthem. To run a phase, read its file and follow it:\n\n- `..\u002F00-scan\u002FSKILL.md` reads `AGENTS.md`\u002F`CLAUDE.md` and prior `state.json`;\n  writes `baseline.json`, `capabilities.json`, `collection.txt`,\n  `hypotheses.json`.\n- `..\u002F01-benchmark\u002FSKILL.md` reads `baseline.json`, `hypotheses.json`; writes\n  `benchmarks.json`, `control.json` and one hash-bound patch per candidate.\n- `..\u002F02-plan\u002FSKILL.md` reads `benchmarks.json`, `baseline.json`; writes\n  `plan.json`.\n- `..\u002F03-execute\u002FSKILL.md` reads `plan.json`, `baseline.json`, `control.json`;\n  writes `execution-log.json` and commits on a new branch in a scratch\n  worktree, leaving the user's checkout, index and branches untouched.\n\n## Step 1: Read-only preflight\n\nWrite no project files here; `00-scan` owns capability probing and creating the\nmemory directory. The runner may update its own environment (below).\n\n1. Read the project's own test and quality-check commands from `AGENTS.md`,\n   `CLAUDE.md`, `justfile`, `tox.ini`, and `pyproject.toml`; the runner\n   (`pytest`, `uv run`, `tox`, `nox`, `hatch`) is the project's choice.\n2. Resolve the memory directory by the lookup order in\n   `..\u002F..\u002Freferences\u002Fmemory-schema.md`. If one exists, take the resume point\n   from `measure.py state resume \u003Cmemory-dir>\u002Fstate.json`.\n3. Read the installed pytest version, for the plan only, through the project's\n   runner so the project environment answers, not the system Python (a runner\n   such as `uv run` may sync its environment first; `uv run --help` lists\n`--no-sync` to avoid it):\n\n   ```console\n   $ pytest --version\n   ```\n\n   Prefix the project's runner when it has one (`uv run pytest --version`,\n   `tox -e py -- --version`).\n\nIf there is no pytest suite, or pytest cannot start for the project's command,\nstop with the halted-run shape in `..\u002F..\u002Freferences\u002Foutput-contract.md`, naming\nthat one cause. A saving is a measured number: when nothing was run, state no\nprojected timing, and call a scope change a hypothesis to benchmark, not a fix.\nA phase whose input is missing stops the same way.\n\n## Orchestration Plan\n\nBefore any measurement or edit, present a plan and get approval.\n\nEnter plan mode before the first edit. In Claude Code call `EnterPlanMode`; in Cursor, Codex, or Gemini use `\u002Fplan` or `Shift+Tab`. Where plan mode is unavailable, present the plan as text.\n\nPresent a plan stating:\n\n- the resolved test and quality-check commands (Step 1);\n- the pytest version, and that `00-scan` will probe plugins and gate\n  capabilities;\n- the resolved memory directory, or that `00-scan` will create it;\n- the starting phase (fresh `00-scan`, or resume) and the scope (`$ARGUMENTS`\n  path or marker, or the whole suite);\n- whether the opt-in typing detectors (`--hygiene`) are on;\n- that suite edits happen only in `03-execute`, inside a scratch worktree on a\n  new branch, one commit per speedup, each applied from its measured patch and\n  verified green with the same collected tests; your checkout is not touched and\n  you merge the branch yourself.\n\nLet the user narrow scope, exclude risky heuristics, or cap the commit count.\nWait for approval, then exit plan mode before running any phase. Without plan\nmode, the phase structure still forces analysis before any write; without an\ninteractive session, stop after presenting the plan.\n\n## Step 2: Run the phases\n\nRead and follow each phase file in order, honoring `--phase` and resume state.\nPass along the test command and memory directory. `00-scan`, `01-benchmark`, and\n`02-plan` never edit the user's working tree. Stop for the user's decision at\neach phase's panel.\n\n## Step 3: Report and decide whether to loop again\n\nAfter `03-execute`, present the result per `..\u002F..\u002Freferences\u002Foutput-contract.md`\nand write it per `..\u002F..\u002Freferences\u002Fprose-rules.md`, then offer a re-scan, the\nopt-in scaffolds, or stop. The contract's closing sentence (the merge command)\ncomes first and carries no note on pushing or on what remains to speed up.\nWithout an interactive question tool, follow the contract's next-step rule.\n\n## Opt-in scaffolds\n\nWhen profiling shows it would help, offer (never impose) a project-owned helper:\n\n- A timing plugin (`..\u002F..\u002Ftemplates\u002Fpytest_optimizer_plugin.py`) that wraps\n  `pytest_fixture_setup` to attribute setup cost to the individual fixture, which\n  pytest core cannot do. It loads from the rootdir `conftest.py` only.\n- A `config.cache` memoization snippet (`..\u002F..\u002Ftemplates\u002Fconftest_cache_snippet.py`)\n  with an invalidation token and an xdist write-guard, for expensive build\n  artifacts recomputed every run. It raises until `INPUTS` is set.\n\nGenerate only on explicit confirmation. The output is untracked; exclude it per\nthe memory-schema reference unless the user chooses to commit it.\n\n## Graceful degradation\n\n- **Old pytest:** gate features via the version matrix; widen `k` when duration\n  precision is coarser (pytest before 6.0 used `time.time()`).\n- **No xdist, randomly, or deadfixtures:** fall back to serial runs, manual\n  reversed-order isolation checks, and the built-in `--fixtures` vs\n  `--fixtures-per-test` diff.\n- **Dirty working tree:** the rule is in `..\u002F..\u002Freferences\u002Fmemory-schema.md`\n  (\"Idempotency token\"). `03-execute` works in its own worktree and does not\n  care about edits made after the scan.\n\n## References\n\n- `..\u002F..\u002Freferences\u002Fscoring-rubric.md`: weighted rubric and the hard safety gate.\n- `..\u002F..\u002Freferences\u002Fheuristic-catalog.md`: heuristic ids with\n  detect\u002Faction\u002Frisk\u002Feffort.\n- `..\u002F..\u002Freferences\u002Fversion-matrix.md`: feature, min version, probe, fallback.\n- `..\u002F..\u002Freferences\u002Fmemory-schema.md`: memory location, file shapes, idempotency\n  token, baseline-diff rule.\n- `..\u002F..\u002Freferences\u002Fdurations-parsing.md`: invoking and parsing `--durations`,\n  phase bucketing, JUnit XML, the noise band.\n- `..\u002F..\u002Fscripts\u002Fmeasure.py`: timing and noise arithmetic, outcome and node-id\n  gates, and the phase files' state, baseline and benchmark records\n  (`measure.py --help`).\n- `..\u002F..\u002Fscripts\u002Fisolate.py`: input digest, staleness check, owned scratch\n  worktrees, bound patches, the commit gate (`isolate.py --help`).\n- `..\u002F..\u002Freferences\u002Ffixture-analysis.md`: scope model and the\n  unused\u002Fduplicate\u002Fmis-scope recipes.\n- `..\u002F..\u002Freferences\u002Fsafety-gates.md`: order independence, collection determinism,\n  xdist preconditions.\n- `..\u002F..\u002Freferences\u002Fparametrize-convention.md`: the typed NamedTuple + test_id\n  convention and typed builtins.\n- `..\u002F..\u002Freferences\u002Foutput-contract.md`: the shared output shape.\n","# pytest-optimizer\n\nProfile a pytest suite, find safe speedups, and apply each as its own verified\ncommit on a new branch, through a resumable, idempotent four-phase loop.\n\nMeasure before you mutate, and prove a speedup against the project's own timing\nnoise before you trust it. A change whose measured saving falls inside the noise\nband, whose run exits non-zero, or whose collected tests change is never\ncommitted, except a deselection the user approved with its test count, which is\nreported as a smaller default run, not a faster suite.\n\n`$ARGUMENTS` may name a test path or marker to scope the run, a `--phase` to jump\nto, `--force` to ignore idempotency tokens and re-run, `--hygiene` to add the opt-in typing\nand parametrize detectors, and `--memory-dir` to\noverride where durable state is written.\n\nPaths under `..\u002F..\u002Fscripts\u002F` resolve from this skill's directory: print the absolute path once and write it literally in each command, since shell variables do not persist between Bash calls.\n\n## The loop\n\n```\n00-scan  ->  01-benchmark  ->  02-plan  ->  03-execute  ->  (re-scan)\n```\n\nEach phase reads and writes durable JSON under the resolved memory directory;\nre-running resumes at the first phase whose `completed` flag is false, which\n`measure.py state resume` reports. After `03-execute`, re-scan and loop until no\ncandidate clears the rubric.\n\nThe phase skills set `disable-model-invocation`, so this skill cannot invoke\nthem. To run a phase, read its file and follow it:\n\n- `..\u002F00-scan\u002FSKILL.md` reads `AGENTS.md`\u002F`CLAUDE.md` and prior `state.json`;\n  writes `baseline.json`, `capabilities.json`, `collection.txt`,\n  `hypotheses.json`.\n- `..\u002F01-benchmark\u002FSKILL.md` reads `baseline.json`, `hypotheses.json`; writes\n  `benchmarks.json`, `control.json` and one hash-bound patch per candidate.\n- `..\u002F02-plan\u002FSKILL.md` reads `benchmarks.json`, `baseline.json`; writes\n  `plan.json`.\n- `..\u002F03-execute\u002FSKILL.md` reads `plan.json`, `baseline.json`, `control.json`;\n  writes `execution-log.json` and commits on a new branch in a scratch\n  worktree, leaving the user's checkout, index and branches untouched.\n\n## Step 1: Read-only preflight\n\nWrite no project files here; `00-scan` owns capability probing and creating the\nmemory directory. The runner may update its own environment (below).\n\n1. Read the project's own test and quality-check commands from `AGENTS.md`,\n   `CLAUDE.md`, `justfile`, `tox.ini`, and `pyproject.toml`; the runner\n   (`pytest`, `uv run`, `tox`, `nox`, `hatch`) is the project's choice.\n2. Resolve the memory directory by the lookup order in\n   `..\u002F..\u002Freferences\u002Fmemory-schema.md`. If one exists, take the resume point\n   from `measure.py state resume \u003Cmemory-dir>\u002Fstate.json`.\n3. Read the installed pytest version, for the plan only, through the project's\n   runner so the project environment answers, not the system Python (a runner\n   such as `uv run` may sync its environment first; `uv run --help` lists\n`--no-sync` to avoid it):\n\n   ```console\n   $ pytest --version\n   ```\n\n   Prefix the project's runner when it has one (`uv run pytest --version`,\n   `tox -e py -- --version`).\n\nIf there is no pytest suite, or pytest cannot start for the project's command,\nstop with the halted-run shape in `..\u002F..\u002Freferences\u002Foutput-contract.md`, naming\nthat one cause. A saving is a measured number: when nothing was run, state no\nprojected timing, and call a scope change a hypothesis to benchmark, not a fix.\nA phase whose input is missing stops the same way.\n\n## Orchestration Plan\n\nBefore any measurement or edit, present a plan and get approval.\n\nEnter plan mode before the first edit. In Claude Code call `EnterPlanMode`; in Cursor, Codex, or Gemini use `\u002Fplan` or `Shift+Tab`. Where plan mode is unavailable, present the plan as text.\n\nPresent a plan stating:\n\n- the resolved test and quality-check commands (Step 1);\n- the pytest version, and that `00-scan` will probe plugins and gate\n  capabilities;\n- the resolved memory directory, or that `00-scan` will create it;\n- the starting phase (fresh `00-scan`, or resume) and the scope (`$ARGUMENTS`\n  path or marker, or the whole suite);\n- whether the opt-in typing detectors (`--hygiene`) are on;\n- that suite edits happen only in `03-execute`, inside a scratch worktree on a\n  new branch, one commit per speedup, each applied from its measured patch and\n  verified green with the same collected tests; your checkout is not touched and\n  you merge the branch yourself.\n\nLet the user narrow scope, exclude risky heuristics, or cap the commit count.\nWait for approval, then exit plan mode before running any phase. Without plan\nmode, the phase structure still forces analysis before any write; without an\ninteractive session, stop after presenting the plan.\n\n## Step 2: Run the phases\n\nRead and follow each phase file in order, honoring `--phase` and resume state.\nPass along the test command and memory directory. `00-scan`, `01-benchmark`, and\n`02-plan` never edit the user's working tree. Stop for the user's decision at\neach phase's panel.\n\n## Step 3: Report and decide whether to loop again\n\nAfter `03-execute`, present the result per `..\u002F..\u002Freferences\u002Foutput-contract.md`\nand write it per `..\u002F..\u002Freferences\u002Fprose-rules.md`, then offer a re-scan, the\nopt-in scaffolds, or stop. The contract's closing sentence (the merge command)\ncomes first and carries no note on pushing or on what remains to speed up.\nWithout an interactive question tool, follow the contract's next-step rule.\n\n## Opt-in scaffolds\n\nWhen profiling shows it would help, offer (never impose) a project-owned helper:\n\n- A timing plugin (`..\u002F..\u002Ftemplates\u002Fpytest_optimizer_plugin.py`) that wraps\n  `pytest_fixture_setup` to attribute setup cost to the individual fixture, which\n  pytest core cannot do. It loads from the rootdir `conftest.py` only.\n- A `config.cache` memoization snippet (`..\u002F..\u002Ftemplates\u002Fconftest_cache_snippet.py`)\n  with an invalidation token and an xdist write-guard, for expensive build\n  artifacts recomputed every run. It raises until `INPUTS` is set.\n\nGenerate only on explicit confirmation. The output is untracked; exclude it per\nthe memory-schema reference unless the user chooses to commit it.\n\n## Graceful degradation\n\n- **Old pytest:** gate features via the version matrix; widen `k` when duration\n  precision is coarser (pytest before 6.0 used `time.time()`).\n- **No xdist, randomly, or deadfixtures:** fall back to serial runs, manual\n  reversed-order isolation checks, and the built-in `--fixtures` vs\n  `--fixtures-per-test` diff.\n- **Dirty working tree:** the rule is in `..\u002F..\u002Freferences\u002Fmemory-schema.md`\n  (\"Idempotency token\"). `03-execute` works in its own worktree and does not\n  care about edits made after the scan.\n\n## References\n\n- `..\u002F..\u002Freferences\u002Fscoring-rubric.md`: weighted rubric and the hard safety gate.\n- `..\u002F..\u002Freferences\u002Fheuristic-catalog.md`: heuristic ids with\n  detect\u002Faction\u002Frisk\u002Feffort.\n- `..\u002F..\u002Freferences\u002Fversion-matrix.md`: feature, min version, probe, fallback.\n- `..\u002F..\u002Freferences\u002Fmemory-schema.md`: memory location, file shapes, idempotency\n  token, baseline-diff rule.\n- `..\u002F..\u002Freferences\u002Fdurations-parsing.md`: invoking and parsing `--durations`,\n  phase bucketing, JUnit XML, the noise band.\n- `..\u002F..\u002Fscripts\u002Fmeasure.py`: timing and noise arithmetic, outcome and node-id\n  gates, and the phase files' state, baseline and benchmark records\n  (`measure.py --help`).\n- `..\u002F..\u002Fscripts\u002Fisolate.py`: input digest, staleness check, owned scratch\n  worktrees, bound patches, the commit gate (`isolate.py --help`).\n- `..\u002F..\u002Freferences\u002Ffixture-analysis.md`: scope model and the\n  unused\u002Fduplicate\u002Fmis-scope recipes.\n- `..\u002F..\u002Freferences\u002Fsafety-gates.md`: order independence, collection determinism,\n  xdist preconditions.\n- `..\u002F..\u002Freferences\u002Fparametrize-convention.md`: the typed NamedTuple + test_id\n  convention and typed builtins.\n- `..\u002F..\u002Freferences\u002Foutput-contract.md`: the shared output shape.\n",{"name":5,"description":7,"allowed-tools":24,"argument-hint":32},[25,26,27,28,29,30,31],"Bash","Read","Grep","Glob","Write","Edit","AskUserQuestion","[test-path-or-marker] [--phase=scan|benchmark|plan|execute] [--force] [--hygiene] [--memory-dir=\u003Cpath>]","main","5fbeb3b47beedea13813be7b7023f736a8d97a65","plugins\u002Fpytest-optimizer\u002Fskills\u002Fpytest-optimizer","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills","https:\u002F\u002Fgithub.com\u002Ftony","https:\u002F\u002Ftony.sh","lighthouse keeper","2026-11-08 17:57:34",{"source":8,"skillName":5,"commands":42,"prompt":52,"zipAvailable":53},[43,46,49],{"tool":44,"command":45},"npx","npx skills add tony\u002Fskills --skill pytest-optimizer",{"tool":47,"command":48},"bunx","bunx skills add tony\u002Fskills --skill pytest-optimizer",{"tool":50,"command":51},"pnpm","pnpm dlx skills add tony\u002Fskills --skill pytest-optimizer","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fpytest-optimizer\u002Fskills\u002Fpytest-optimizer\u002FSKILL.md",true,[55],{"path":56,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":59,"sourceHtmlUrl":60},"agents\u002Fopenai.yaml","REFERENCE","参考文档",255,"https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fpytest-optimizer\u002Fskills\u002Fpytest-optimizer\u002Fagents\u002Fopenai.yaml",[62,69,76,83,90,97],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7E8NCSHXV2JPW4J61R0TQS7","serene-bliss","Use when asked to make a CLI --help, docs page, error message, or tooling UX more serene or calmer for developer experience; returns text, no edits.","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fweave\u002Fskills\u002Fserene-bliss\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7E8N87MAVJD8RRQKQB1ZPS6","review","Weave code review: run independent adversarial reviews in parallel, then synthesize findings","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fweave\u002Fskills\u002Freview\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7E8N402XG4ZBFNACQ413163","refine","Use when asked to refine a draft, iterate on code until cleaner, or polish a README through repeated critique and improvement.","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fweave\u002Fskills\u002Frefine\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7E8N01KB86A7136P00GVKJW","prompt","Weave prompt: compare independent adversarial implementations, then adopt the single best one","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fweave\u002Fskills\u002Fprompt\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7E8MW7YJW2PCMDW3G9RTGNK","plan","Weave planning: compare independent adversarial plans, then synthesize the best one","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fweave\u002Fskills\u002Fplan\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":19},"01M7E8MSBYXNCV4ZYWDS4PFJXK","fix-review","Fix weave review findings: validate, add test coverage, fix, and commit each as atomic changes","https:\u002F\u002Fgithub.com\u002Ftony\u002Fskills\u002Fblob\u002Fmain\u002Fplugins\u002Fweave\u002Fskills\u002Ffix-review\u002FSKILL.md",[],[]]