[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M753DY3MCEV0ADSS408F2B4E":3,"skill-same:01M753DY3MCEV0ADSS408F2B4E":66},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M753DY3MCEV0ADSS408F2B4E","refresh-repo-skill",null,"Refreshes an existing repository-specific Agent Skill after the source repository changed. Use when the user says repo code, APIs, docs, examples, configs, dependencies, or behavior changed and an old skill may now be stale, outdated, inconsistent with current code, or needs to be resynchronized from repository evidence.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fcli\u002Fpackages\u002Fcoding-agent\u002Fsrc\u002Fdisco\u002Fskills\u002Frefresh-repo-skill\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: refresh-repo-skill\ndescription: \"Refreshes an existing repository-specific Agent Skill after the source repository changed. Use when the user says repo code, APIs, docs, examples, configs, dependencies, or behavior changed and an old skill may now be stale, outdated, inconsistent with current code, or needs to be resynchronized from repository evidence.\"\nmetadata:\n  disco-role: meta\n---\n\n# Refresh Repo Skill\n\n## Purpose\n\nUse this skill when an existing repo-specific Agent Skill should continue\ndescribing the same repository, but the repository itself has changed since the\nskill was created or last maintained.\n\nTypical requests include:\n\n- Refresh an old skill after new commits, releases, branches, APIs, CLIs,\n  config formats, examples, or dependencies changed.\n- Audit a skill for stale claims against the current repository code.\n- Update copied references, scripts, workflows, troubleshooting, or usability\n  cases that no longer match the repo.\n- Rebuild the skill's repository evidence without discarding the skill's useful\n  identity, routing, or prior coverage.\n\nThis workflow is different from `extend-repo-skill`: use\n`extend-repo-skill` when the user asks to add a new capability or deeper\ncoverage to a working skill. Use this skill when repository drift is the main\nreason the skill may be wrong.\n\n## Inputs\n\nGather or infer:\n\n- Existing skill directory containing `SKILL.md`.\n- Current repository path to use as the source of truth.\n- Existing `references\u002Frepo-provenance.md` from the skill, when present.\n- Optional previous repository baseline, branch, tag, release, commit, or date\n  when the skill has no provenance file.\n- Python inspection environment and installed package name when live API or CLI\n  verification is needed.\n- Existing review\u002Ftest artifact directory, if any.\n- Desired review\u002Ftest artifact output directory, if the user has a preference.\n\nIf the existing skill directory is missing or does not contain `SKILL.md`, stop\nand ask for the correct skill path. If the repository path is missing and cannot\nbe inferred from the current working directory or user request, ask for it.\n\n## Reference Map\n\nRead these references as the workflow reaches each stage:\n\n- [references\u002Fchange-detection-and-staleness-audit.md](references\u002Fchange-detection-and-staleness-audit.md):\n  identify the current repo state, optional baseline, stale claims, and affected\n  skill areas.\n- [references\u002Frefresh-editing.md](references\u002Frefresh-editing.md): edit the\n  existing skill in place, replace outdated guidance, preserve useful identity,\n  and update tests\u002Fscripts\u002Freferences.\n- [references\u002Fverification-and-handoff.md](references\u002Fverification-and-handoff.md):\n  run stale-claim checks, live verification, usability review, and final\n  handoff.\n- [..\u002Fverify-repo-skill\u002Fscripts\u002Fresolve_repo_license.mjs](..\u002Fverify-repo-skill\u002Fscripts\u002Fresolve_repo_license.mjs)\n  and [..\u002Fverify-repo-skill\u002Fscripts\u002Fapply_repo_license.mjs](..\u002Fverify-repo-skill\u002Fscripts\u002Fapply_repo_license.mjs):\n  resolve and recursively apply the source repository license for the exact\n  refresh commit before final verification.\n\nUse [scripts\u002Fcheck_repo_provenance.py](scripts\u002Fcheck_repo_provenance.py) when\n`references\u002Frepo-provenance.md` exists. It compares the skill snapshot with the\ncurrent Git checkout and prints JSON with `current`, `stale`, or `unknown`\nstatus.\n\nWhen useful, also read sibling workflow-skill references:\n\n- `..\u002Fcreate-repo-skill\u002Freferences\u002F` for repository evidence discovery,\n  installed-package inspection, and output structure.\n- `..\u002Fverify-repo-skill\u002Freferences\u002F` for usability case format,\n  verification review, and import\u002Findex routing guidance.\n- `..\u002Fextend-repo-skill\u002Freferences\u002F` for narrow editing and regression\n  review rules.\n- `..\u002Fprepare-repo-skill-env\u002FSKILL.md` if live Python\n  inspection is needed and no verified environment is available.\n\n## Required Workflow\n\n1. Resolve the existing skill directory, current repository path, existing\n   `references\u002Frepo-provenance.md` baseline, optional user-provided baseline,\n   Python inspection context, review\u002Ftest artifact directory, and review output\n   path. If the user does not specify an artifact directory, default to\n   `\u003Crepository-path>\u002Fskills\u002Ftests\u002F\u003Cskill-id>\u002F`, with usability cases under\n   `test-cases\u002F` and review reports under `reports\u002F`. If the existing skill is\n   the live managed copy under `\u003Cagent-dir>\u002Fskills\u002Frepositories\u002Frepo-skills\u002F\u003Cskill-id>\u002F`, copy\n   only its runtime tree to a temporary working directory outside\n   `\u003Cagent-dir>\u002Fskills\u002F` before editing. Keep that working copy until the\n   dedicated importer succeeds; never edit the live managed copy in place.\n2. Read [references\u002Fchange-detection-and-staleness-audit.md](references\u002Fchange-detection-and-staleness-audit.md).\n   Build a current-state map of the existing skill. When provenance exists, run\n   `python scripts\u002Fcheck_repo_provenance.py --skill-dir \u003Cskill_dir> --repo-path \u003Crepo_path>`\n   from this workflow-skill directory or adapt the script path to the installed\n   skill copy. Use its JSON output as the first staleness signal, then gather\n   current repository evidence and optional change evidence from Git history or\n   release notes.\n3. Produce a staleness audit that separates:\n   - Claims still supported by current repo evidence.\n   - Claims that are stale, removed, renamed, or behaviorally changed.\n   - New repo capabilities that should be represented because they replace or\n     materially alter existing skill guidance.\n   - Unknowns requiring live inspection or user clarification.\n4. If live Python inspection is required and no verified environment exists, use\n   `prepare-repo-skill-env` to create or repair one, then\n   continue only from its verified handoff.\n5. Read [references\u002Frefresh-editing.md](references\u002Frefresh-editing.md). Edit the\n   resolved runtime working copy in place. Preserve root and sub-skill identities unless the\n   user explicitly asks for a rename or the current name is invalid. Add or\n   update `references\u002Frepo-provenance.md` with the refreshed source snapshot.\n   Preserve the canonical `repo_id` and `skill_id` by default. Treat the\n   existing area-family assignments as the routing baseline: keep them when\n   the repository's capability scope is unchanged, but do not keep them\n   blindly when the refresh adds, removes, or materially changes a capability.\n   If the capability scope changed, the taxonomy hash changed, or the user\n   requests reclassification, produce a new external area-family routing\n   handoff and matching minimal v2 metadata before import. Do not hand-edit\n   generated router Markdown or silently change assignments in prose.\n6. Resolve and apply the repository license after the refreshed source commit\n   is known and before handing the tree to verification:\n\n   ```bash\n   node ..\u002Fverify-repo-skill\u002Fscripts\u002Fresolve_repo_license.mjs \\\n     --repository \u003Cowner\u002Frepository> \\\n     --source-commit \u003C40-hex-source-commit> \\\n     --json > \u003Cartifact-root>\u002Freports\u002Flicense-resolution.json\n   node ..\u002Fverify-repo-skill\u002Fscripts\u002Fapply_repo_license.mjs \\\n     --skill-dir \u003Cexternal-runtime-skill-dir> \\\n     --license \u003Cvalue-from-report>\n   ```\n\n   Re-query on every refresh; do not preserve a stale license solely because it\n   was present in the previous skill. Update all root\u002Fsub-skill frontmatter to\n   the same result. Preserve `NOASSERTION` when GitHub returns it; normalize only\n   unavailable results to `NO_LICENSE`, keep the report outside the runtime tree,\n   and carry old value, new value, source commit, status, and reason into the\n   handoff. `NO_LICENSE` is a warning and not a legal conclusion.\n7. Update usability test cases under the review\u002Ftest artifact directory's\n   `test-cases\u002F` subtree so at least one case proves refreshed behavior and at\n   least one case guards a pre-existing workflow that should remain valid.\n8. Read [references\u002Fverification-and-handoff.md](references\u002Fverification-and-handoff.md).\n   Verify that refreshed public skill content is self-contained, current,\n   privacy-safe, and reachable from nearby `SKILL.md` files. Save staleness\n   audits, verification reports, and human-review notes under the review\u002Ftest\n   artifact directory's `reports\u002F` subtree.\n9. After verification passes, follow `verify-repo-skill`'s structured import\n   policy: use `ask_user_question` when approval is still required, then run the\n   approved or auto-authorized refresh through\n   `verify-repo-skill\u002Fscripts\u002Fimport_repo_skill.mjs` with `--overwrite`, the\n   refreshed runtime skill directory, and `--routing-entry\n   \u003Crepo-path>\u002Fskills\u002Fdisco\u002Frouting_decision\u002Fclassification.json`. The external\n   handoff is mandatory for a normal classified import, even when the existing\n   area-family assignments are retained. Use `--overwrite` only because this\n   workflow is updating that exact approved repo skill. The helper installs the\n   refreshed runtime tree under\n   `~\u002F.disco\u002Fagent\u002Fskills\u002Frepositories\u002Frepo-skills\u002F\u003Cskill-id>\u002F`, rebuilds the\n   sibling live DisCo `repo-skills-router` under the global lock, and restores\n   both on failure. Do not hand-edit router Markdown or manually combine copy\n   and updater commands. The refreshed skill is directly available to DisCo\n   Researcher in a new session; use `import-repo-skills-to-agent` only when the\n   user explicitly asks for a cross-agent export.\n\n## Non-Negotiables\n\n- Treat the current repository as the source of truth; do not preserve old skill\n  claims just because they were previously useful.\n- Do not regenerate a replacement skill from scratch unless the user explicitly\n  asks or the old skill is so structurally invalid that in-place repair would be\n  misleading.\n- Do not edit a live skill under `\u003Cagent-dir>\u002Fskills\u002Frepositories\u002Frepo-skills\u002F` before the\n  global import lock is held. Refresh an external working copy, then let\n  `verify-repo-skill\u002Fscripts\u002Fimport_repo_skill.mjs` replace the live target.\n- Do not finish a refresh without `references\u002Frepo-provenance.md`; if the old\n  skill lacks it, create it from the current repository snapshot.\n- Do not silently change a repository's `repo_id` or `skill_id` during a\n  refresh. A rename is a separate identity migration and must update the\n  central repository index and all router references together.\n- Do not silently change area-family assignments. Preserve the prior routing\n  when capability scope is unchanged; otherwise use the verified external\n  routing handoff and minimal v2 metadata, then let the locked importer update\n  the router and indexes atomically.\n- Do not change the skill's public purpose to cover unrelated new repository\n  areas unless those areas replace or materially affect existing guidance.\n- Do not add claims about APIs, CLIs, configs, data formats, dependency\n  behavior, or runtime behavior without current repository evidence or live\n  inspection.\n- Do not leak local checkout paths, Python executable paths, conda or virtualenv\n  names, `pip show` locations, API keys, or machine-specific details into public\n  skill files.\n- Do not link runtime skill documentation to the source repository checkout.\n  Distill current facts into the skill directory.\n- Do not preserve or reintroduce runtime instructions that refer to source repo\n  scripts, examples, notebooks, tools, or configs by path. If the refreshed\n  skill still needs that behavior, add a bundled replacement under the skill's\n  own `scripts\u002F` or `references\u002F` tree and link it instead.\n- Do not treat `tests\u002F` as a skill directory. It is the review\u002Ftest artifact area.\n- Do not write staleness audits, `evals\u002F`, verification reports, human-review\n  notes, publication checklists, prompt samples, benchmark notes, or other\n  check-only artifacts inside the runtime skill directory. Put concrete\n  usability cases under the review\u002Ftest artifact directory's `test-cases\u002F`\n  subtree and reports under its `reports\u002F` subtree, defaulting to\n  `\u003Crepository-path>\u002Fskills\u002Ftests\u002F\u003Cskill-id>\u002F`.\n\n## Output Summary\n\nBy the end, the user should have:\n\n- A runtime working copy refreshed against current repository evidence, followed\n  by a locked live replacement when import is approved or auto-authorized.\n- An updated `references\u002Frepo-provenance.md` snapshot that future agents can use\n  to detect whether the skill may be stale.\n- A staleness audit and review package under the review\u002Ftest artifact\n  directory's `reports\u002F` subtree that identifies changed, retained, and removed\n  guidance.\n- Updated references, scripts, sub-skills, and usability test cases where the\n  old skill was stale.\n- Verification evidence showing the public skill no longer depends on outdated\n  repo facts.\n- A routing handoff stating whether prior area-family assignments were\n  preserved or reclassified, with the reason and evidence for any change.\n- A final handoff that distinguishes refreshed public skill content, review\u002Ftest\n  artifacts, evidence used, and any accepted uncertainty.\n","# Refresh Repo Skill\n\n## Purpose\n\nUse this skill when an existing repo-specific Agent Skill should continue\ndescribing the same repository, but the repository itself has changed since the\nskill was created or last maintained.\n\nTypical requests include:\n\n- Refresh an old skill after new commits, releases, branches, APIs, CLIs,\n  config formats, examples, or dependencies changed.\n- Audit a skill for stale claims against the current repository code.\n- Update copied references, scripts, workflows, troubleshooting, or usability\n  cases that no longer match the repo.\n- Rebuild the skill's repository evidence without discarding the skill's useful\n  identity, routing, or prior coverage.\n\nThis workflow is different from `extend-repo-skill`: use\n`extend-repo-skill` when the user asks to add a new capability or deeper\ncoverage to a working skill. Use this skill when repository drift is the main\nreason the skill may be wrong.\n\n## Inputs\n\nGather or infer:\n\n- Existing skill directory containing `SKILL.md`.\n- Current repository path to use as the source of truth.\n- Existing `references\u002Frepo-provenance.md` from the skill, when present.\n- Optional previous repository baseline, branch, tag, release, commit, or date\n  when the skill has no provenance file.\n- Python inspection environment and installed package name when live API or CLI\n  verification is needed.\n- Existing review\u002Ftest artifact directory, if any.\n- Desired review\u002Ftest artifact output directory, if the user has a preference.\n\nIf the existing skill directory is missing or does not contain `SKILL.md`, stop\nand ask for the correct skill path. If the repository path is missing and cannot\nbe inferred from the current working directory or user request, ask for it.\n\n## Reference Map\n\nRead these references as the workflow reaches each stage:\n\n- [references\u002Fchange-detection-and-staleness-audit.md](references\u002Fchange-detection-and-staleness-audit.md):\n  identify the current repo state, optional baseline, stale claims, and affected\n  skill areas.\n- [references\u002Frefresh-editing.md](references\u002Frefresh-editing.md): edit the\n  existing skill in place, replace outdated guidance, preserve useful identity,\n  and update tests\u002Fscripts\u002Freferences.\n- [references\u002Fverification-and-handoff.md](references\u002Fverification-and-handoff.md):\n  run stale-claim checks, live verification, usability review, and final\n  handoff.\n- [..\u002Fverify-repo-skill\u002Fscripts\u002Fresolve_repo_license.mjs](..\u002Fverify-repo-skill\u002Fscripts\u002Fresolve_repo_license.mjs)\n  and [..\u002Fverify-repo-skill\u002Fscripts\u002Fapply_repo_license.mjs](..\u002Fverify-repo-skill\u002Fscripts\u002Fapply_repo_license.mjs):\n  resolve and recursively apply the source repository license for the exact\n  refresh commit before final verification.\n\nUse [scripts\u002Fcheck_repo_provenance.py](scripts\u002Fcheck_repo_provenance.py) when\n`references\u002Frepo-provenance.md` exists. It compares the skill snapshot with the\ncurrent Git checkout and prints JSON with `current`, `stale`, or `unknown`\nstatus.\n\nWhen useful, also read sibling workflow-skill references:\n\n- `..\u002Fcreate-repo-skill\u002Freferences\u002F` for repository evidence discovery,\n  installed-package inspection, and output structure.\n- `..\u002Fverify-repo-skill\u002Freferences\u002F` for usability case format,\n  verification review, and import\u002Findex routing guidance.\n- `..\u002Fextend-repo-skill\u002Freferences\u002F` for narrow editing and regression\n  review rules.\n- `..\u002Fprepare-repo-skill-env\u002FSKILL.md` if live Python\n  inspection is needed and no verified environment is available.\n\n## Required Workflow\n\n1. Resolve the existing skill directory, current repository path, existing\n   `references\u002Frepo-provenance.md` baseline, optional user-provided baseline,\n   Python inspection context, review\u002Ftest artifact directory, and review output\n   path. If the user does not specify an artifact directory, default to\n   `\u003Crepository-path>\u002Fskills\u002Ftests\u002F\u003Cskill-id>\u002F`, with usability cases under\n   `test-cases\u002F` and review reports under `reports\u002F`. If the existing skill is\n   the live managed copy under `\u003Cagent-dir>\u002Fskills\u002Frepositories\u002Frepo-skills\u002F\u003Cskill-id>\u002F`, copy\n   only its runtime tree to a temporary working directory outside\n   `\u003Cagent-dir>\u002Fskills\u002F` before editing. Keep that working copy until the\n   dedicated importer succeeds; never edit the live managed copy in place.\n2. Read [references\u002Fchange-detection-and-staleness-audit.md](references\u002Fchange-detection-and-staleness-audit.md).\n   Build a current-state map of the existing skill. When provenance exists, run\n   `python scripts\u002Fcheck_repo_provenance.py --skill-dir \u003Cskill_dir> --repo-path \u003Crepo_path>`\n   from this workflow-skill directory or adapt the script path to the installed\n   skill copy. Use its JSON output as the first staleness signal, then gather\n   current repository evidence and optional change evidence from Git history or\n   release notes.\n3. Produce a staleness audit that separates:\n   - Claims still supported by current repo evidence.\n   - Claims that are stale, removed, renamed, or behaviorally changed.\n   - New repo capabilities that should be represented because they replace or\n     materially alter existing skill guidance.\n   - Unknowns requiring live inspection or user clarification.\n4. If live Python inspection is required and no verified environment exists, use\n   `prepare-repo-skill-env` to create or repair one, then\n   continue only from its verified handoff.\n5. Read [references\u002Frefresh-editing.md](references\u002Frefresh-editing.md). Edit the\n   resolved runtime working copy in place. Preserve root and sub-skill identities unless the\n   user explicitly asks for a rename or the current name is invalid. Add or\n   update `references\u002Frepo-provenance.md` with the refreshed source snapshot.\n   Preserve the canonical `repo_id` and `skill_id` by default. Treat the\n   existing area-family assignments as the routing baseline: keep them when\n   the repository's capability scope is unchanged, but do not keep them\n   blindly when the refresh adds, removes, or materially changes a capability.\n   If the capability scope changed, the taxonomy hash changed, or the user\n   requests reclassification, produce a new external area-family routing\n   handoff and matching minimal v2 metadata before import. Do not hand-edit\n   generated router Markdown or silently change assignments in prose.\n6. Resolve and apply the repository license after the refreshed source commit\n   is known and before handing the tree to verification:\n\n   ```bash\n   node ..\u002Fverify-repo-skill\u002Fscripts\u002Fresolve_repo_license.mjs \\\n     --repository \u003Cowner\u002Frepository> \\\n     --source-commit \u003C40-hex-source-commit> \\\n     --json > \u003Cartifact-root>\u002Freports\u002Flicense-resolution.json\n   node ..\u002Fverify-repo-skill\u002Fscripts\u002Fapply_repo_license.mjs \\\n     --skill-dir \u003Cexternal-runtime-skill-dir> \\\n     --license \u003Cvalue-from-report>\n   ```\n\n   Re-query on every refresh; do not preserve a stale license solely because it\n   was present in the previous skill. Update all root\u002Fsub-skill frontmatter to\n   the same result. Preserve `NOASSERTION` when GitHub returns it; normalize only\n   unavailable results to `NO_LICENSE`, keep the report outside the runtime tree,\n   and carry old value, new value, source commit, status, and reason into the\n   handoff. `NO_LICENSE` is a warning and not a legal conclusion.\n7. Update usability test cases under the review\u002Ftest artifact directory's\n   `test-cases\u002F` subtree so at least one case proves refreshed behavior and at\n   least one case guards a pre-existing workflow that should remain valid.\n8. Read [references\u002Fverification-and-handoff.md](references\u002Fverification-and-handoff.md).\n   Verify that refreshed public skill content is self-contained, current,\n   privacy-safe, and reachable from nearby `SKILL.md` files. Save staleness\n   audits, verification reports, and human-review notes under the review\u002Ftest\n   artifact directory's `reports\u002F` subtree.\n9. After verification passes, follow `verify-repo-skill`'s structured import\n   policy: use `ask_user_question` when approval is still required, then run the\n   approved or auto-authorized refresh through\n   `verify-repo-skill\u002Fscripts\u002Fimport_repo_skill.mjs` with `--overwrite`, the\n   refreshed runtime skill directory, and `--routing-entry\n   \u003Crepo-path>\u002Fskills\u002Fdisco\u002Frouting_decision\u002Fclassification.json`. The external\n   handoff is mandatory for a normal classified import, even when the existing\n   area-family assignments are retained. Use `--overwrite` only because this\n   workflow is updating that exact approved repo skill. The helper installs the\n   refreshed runtime tree under\n   `~\u002F.disco\u002Fagent\u002Fskills\u002Frepositories\u002Frepo-skills\u002F\u003Cskill-id>\u002F`, rebuilds the\n   sibling live DisCo `repo-skills-router` under the global lock, and restores\n   both on failure. Do not hand-edit router Markdown or manually combine copy\n   and updater commands. The refreshed skill is directly available to DisCo\n   Researcher in a new session; use `import-repo-skills-to-agent` only when the\n   user explicitly asks for a cross-agent export.\n\n## Non-Negotiables\n\n- Treat the current repository as the source of truth; do not preserve old skill\n  claims just because they were previously useful.\n- Do not regenerate a replacement skill from scratch unless the user explicitly\n  asks or the old skill is so structurally invalid that in-place repair would be\n  misleading.\n- Do not edit a live skill under `\u003Cagent-dir>\u002Fskills\u002Frepositories\u002Frepo-skills\u002F` before the\n  global import lock is held. Refresh an external working copy, then let\n  `verify-repo-skill\u002Fscripts\u002Fimport_repo_skill.mjs` replace the live target.\n- Do not finish a refresh without `references\u002Frepo-provenance.md`; if the old\n  skill lacks it, create it from the current repository snapshot.\n- Do not silently change a repository's `repo_id` or `skill_id` during a\n  refresh. A rename is a separate identity migration and must update the\n  central repository index and all router references together.\n- Do not silently change area-family assignments. Preserve the prior routing\n  when capability scope is unchanged; otherwise use the verified external\n  routing handoff and minimal v2 metadata, then let the locked importer update\n  the router and indexes atomically.\n- Do not change the skill's public purpose to cover unrelated new repository\n  areas unless those areas replace or materially affect existing guidance.\n- Do not add claims about APIs, CLIs, configs, data formats, dependency\n  behavior, or runtime behavior without current repository evidence or live\n  inspection.\n- Do not leak local checkout paths, Python executable paths, conda or virtualenv\n  names, `pip show` locations, API keys, or machine-specific details into public\n  skill files.\n- Do not link runtime skill documentation to the source repository checkout.\n  Distill current facts into the skill directory.\n- Do not preserve or reintroduce runtime instructions that refer to source repo\n  scripts, examples, notebooks, tools, or configs by path. If the refreshed\n  skill still needs that behavior, add a bundled replacement under the skill's\n  own `scripts\u002F` or `references\u002F` tree and link it instead.\n- Do not treat `tests\u002F` as a skill directory. It is the review\u002Ftest artifact area.\n- Do not write staleness audits, `evals\u002F`, verification reports, human-review\n  notes, publication checklists, prompt samples, benchmark notes, or other\n  check-only artifacts inside the runtime skill directory. Put concrete\n  usability cases under the review\u002Ftest artifact directory's `test-cases\u002F`\n  subtree and reports under its `reports\u002F` subtree, defaulting to\n  `\u003Crepository-path>\u002Fskills\u002Ftests\u002F\u003Cskill-id>\u002F`.\n\n## Output Summary\n\nBy the end, the user should have:\n\n- A runtime working copy refreshed against current repository evidence, followed\n  by a locked live replacement when import is approved or auto-authorized.\n- An updated `references\u002Frepo-provenance.md` snapshot that future agents can use\n  to detect whether the skill may be stale.\n- A staleness audit and review package under the review\u002Ftest artifact\n  directory's `reports\u002F` subtree that identifies changed, retained, and removed\n  guidance.\n- Updated references, scripts, sub-skills, and usability test cases where the\n  old skill was stale.\n- Verification evidence showing the public skill no longer depends on outdated\n  repo facts.\n- A routing handoff stating whether prior area-family assignments were\n  preserved or reclassified, with the reason and evidence for any change.\n- A final handoff that distinguishes refreshed public skill content, review\u002Ftest\n  artifacts, evidence used, and any accepted uncertainty.\n",{"name":5,"metadata":23,"description":7},{"disco-role":24},"meta","main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","cli\u002Fpackages\u002Fcoding-agent\u002Fsrc\u002Fdisco\u002Fskills\u002Frefresh-repo-skill",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill refresh-repo-skill",{"tool":38,"command":39},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill refresh-repo-skill",{"tool":41,"command":42},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill refresh-repo-skill","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fcli\u002Fpackages\u002Fcoding-agent\u002Fsrc\u002Fdisco\u002Fskills\u002Frefresh-repo-skill\u002FSKILL.md",true,[46,52,56,60],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fchange-detection-and-staleness-audit.md","REFERENCE","参考文档",8253,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fcli\u002Fpackages\u002Fcoding-agent\u002Fsrc\u002Fdisco\u002Fskills\u002Frefresh-repo-skill\u002Freferences\u002Fchange-detection-and-staleness-audit.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Frefresh-editing.md",7975,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fcli\u002Fpackages\u002Fcoding-agent\u002Fsrc\u002Fdisco\u002Fskills\u002Frefresh-repo-skill\u002Freferences\u002Frefresh-editing.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fverification-and-handoff.md",9027,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fcli\u002Fpackages\u002Fcoding-agent\u002Fsrc\u002Fdisco\u002Fskills\u002Frefresh-repo-skill\u002Freferences\u002Fverification-and-handoff.md",{"path":61,"resourceType":62,"resourceTypeDesc":63,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Fcheck_repo_provenance.py","SCRIPT","脚本",6278,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fcli\u002Fpackages\u002Fcoding-agent\u002Fsrc\u002Fdisco\u002Fskills\u002Frefresh-repo-skill\u002Fscripts\u002Fcheck_repo_provenance.py",[67,75,82,89,96,103],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":73,"publishedAt":74},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":74},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":74},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":74},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":74},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":74},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]