[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CVR3VBVQ7GCR6C2WD1BEZS":3,"skill-same:01M6CVR3VBVQ7GCR6C2WD1BEZS":84},{"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":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":44,"commitSha":45,"skillDirectory":46,"forksCount":47,"repositoryHtmlUrl":48,"repositoryHomepageUrl":6,"ownerHtmlUrl":49,"ownerBlogUrl":50,"ownerDescription":6,"updatedAt":51,"install":52,"resources":65},"01M6CVR3VBVQ7GCR6C2WD1BEZS","lifecycle-audit",null,"Run a black-box Mycelium lifecycle smoke test through a real Claude Code or Codex CLI task. Use after plugin installation, hook, SessionStart, PostToolUse, Stop, activity tracking, handoff, or data-lineage changes; when automatic hooks appear not to fire; or before merging cross-host lifecycle work. Verify installed-build identity, natural dispatch, state transitions, positive and negative probes, and scientific-tree isolation. Do not use as a substitute for unit tests or for ordinary scientific analysis.","arjunrajlaboratory\u002Fmycelium","arjunrajlaboratory","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F59969125?v=4","01M6CVPTM66Z3J8E20GYXJ1RQJ","Arjun Raj's systems biology lab","01M6CVPTRFA9TWRGGB81YN4AP0",72,"MIT","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Flifecycle-audit\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"lifecycle","lifecycle audit","audit","analysis","codex","task","do","smoke test","hooks","smoke","use","work","unit tests","cli","test","lineage","state","code","plugin","codex cli","2026-11-10 10:12:30",false,"---\nname: lifecycle-audit\ndescription: >\n  Run a black-box Mycelium lifecycle smoke test through a real Claude Code or\n  Codex CLI task. Use after plugin installation, hook, SessionStart,\n  PostToolUse, Stop, activity tracking, handoff, or data-lineage changes; when\n  automatic hooks appear not to fire; or before merging cross-host lifecycle\n  work. Verify installed-build identity, natural dispatch, state transitions,\n  positive and negative probes, and scientific-tree isolation. Do not use as a\n  substitute for unit tests or for ordinary scientific analysis.\n---\n\n# Mycelium Lifecycle Audit\n\nResolve bundled paths relative to the Mycelium plugin root, two directories\nabove this `SKILL.md`. Read `references\u002Faudit-protocol.md` completely before\nlaunching a host process.\n\nThis is a black-box integration audit. A manually invoked hook can diagnose hook\nlogic, but it cannot prove that Claude Code or Codex discovered, trusted, and\ndispatched the hook.\n\n## 1. Fix the audit boundary\n\n- Record the host, target repository, source commit, installed plugin version,\n  and expected plugin root.\n- Use a disposable initialized repository when possible. If using a real\n  analytical project, capture a baseline and prohibit scientific code, data,\n  output, report, and pre-existing `.living` edits.\n- Confirm the requested CLI is installed. Inspect its current help rather than\n  guessing flags.\n- Freeze the source checkout and installed artifact for the entire host run.\n  Do not edit, reinstall, or replace an exercised hook while its process may be\n  reading it.\n- If the user did not explicitly request a real host run, confirm before sending\n  repository context to an external agent service.\n\nFor Codex local development, verify `codex plugin list --json`, use the supported\ncachebuster\u002Freinstall flow, and start a new task. Hash exercised installed files\nagainst the source candidate. For Claude, identify the plugin source actually\nloaded by the CLI and verify its files the same way. A source\u002Fcache mismatch is\nan audit failure even if hooks otherwise work.\n\n## 2. Capture a pre-run baseline\n\nRecord:\n\n- `git status` and relevant file hashes;\n- existing session log, registry, handoff, lineage, and runtime markers;\n- scientific files and output state;\n- expected next log or session identifier when determinable.\n\nDo not delete suspicious state merely to make the test pass. If a clean-room\naudit is needed, create a fresh disposable repository.\n\n## 3. Launch one genuinely fresh host task\n\nRun the host CLI from the exact target root. Do not invoke any Mycelium hook\nscript in the task or immediately afterward. Use the canonical child-task\nprompt in `references\u002Fchild-task-prompt.md` — every probe it carries is\nrequired, including the nested-subagent probe, which may be marked N\u002FA only\nwhen the host exposes no native subagent mechanism. Ask the agent to expose\nthe exact automatic lifecycle context it receives and perform only the probes\nin the audit protocol. The launcher, not the child agent, owns artifact verification\nand baseline capture. Where the CLI supports it, disable skills in the child\ntask so the lifecycle-audit skill does not recursively launch or expand itself.\nKeep ordinary file-reading and editing tools available: the child needs them to\nobserve state and satisfy the expected Stop-compliance retry.\n\nExercise:\n\n1. SessionStart injection and active-state creation.\n2. When the host exposes a native subagent\u002Ftask mechanism, one read-only child\n   start and Stop while the primary remains active.\n3. A known-positive harmless analysis command.\n4. A near-neighbor negative command that must remain silent.\n5. One disposable successful edit using the host's normal edit tool. Remove it\n   with that tool when deletion is supported; otherwise, after edit-tracking\n   evidence is captured, use the protocol's single explicit shell cleanup.\n6. Natural Stop, including the expected enforcement block, compliant retry,\n   and a complete five-section handoff using Mycelium's exact headings once\n   each, in order, with nonblank section bodies.\n\nUse `python3` unless the target explicitly provides another interpreter; do not\nassume `python` exists. Preserve the real command exit status—an optional\ndependency failure may still be a valid lineage probe.\n\n## 4. Inspect only after natural Stop\n\nCompare post-run state with the baseline. Verify the first Stop blocked until\nthe disposable repository's living layer was updated, then verify exact-once finalization,\nregistry and lineage effects, handoff structure, cleanup, and absence of the\ndisposable file. Separate lifecycle-owned changes from agent-authored scientific\nchanges.\n\nClassify every unexpected result as one of:\n\n- **Artifact identity:** wrong source, version, cache, or packaged file.\n- **Host dispatch:** registration, trust, discovery, matcher, or host payload\n  never reached the hook.\n- **Hook logic:** dispatch occurred, but state or context violated the contract.\n- **Environment:** missing interpreter\u002Fdependency or CLI limitation unrelated to\n  Mycelium behavior.\n- **Agent compliance:** hook instructed the agent correctly, but it ignored or\n  rewrote required state.\n- **Inconclusive:** evidence was overwritten, manually stimulated, or never\n  captured.\n\nNever relabel an environment or compliance failure as a hook pass without\nshowing the hook evidence that did pass.\n\n## 5. Report and clean up\n\nReturn the evidence table from the audit protocol, exact errors, plugin identity,\nand a scientific-state diff. Remove only disposable probe files you created;\nretain failing lifecycle state when it is needed to reproduce a retry bug.\n\nDo not repair failures during the audit. Diagnose first so the observed state is\nnot contaminated. If a fix is requested afterward, switch to\n`$mycelium:develop`, add a red regression test, and repeat the black-box audit on\nthe new installed artifact.\n","# Mycelium Lifecycle Audit\n\nResolve bundled paths relative to the Mycelium plugin root, two directories\nabove this `SKILL.md`. Read `references\u002Faudit-protocol.md` completely before\nlaunching a host process.\n\nThis is a black-box integration audit. A manually invoked hook can diagnose hook\nlogic, but it cannot prove that Claude Code or Codex discovered, trusted, and\ndispatched the hook.\n\n## 1. Fix the audit boundary\n\n- Record the host, target repository, source commit, installed plugin version,\n  and expected plugin root.\n- Use a disposable initialized repository when possible. If using a real\n  analytical project, capture a baseline and prohibit scientific code, data,\n  output, report, and pre-existing `.living` edits.\n- Confirm the requested CLI is installed. Inspect its current help rather than\n  guessing flags.\n- Freeze the source checkout and installed artifact for the entire host run.\n  Do not edit, reinstall, or replace an exercised hook while its process may be\n  reading it.\n- If the user did not explicitly request a real host run, confirm before sending\n  repository context to an external agent service.\n\nFor Codex local development, verify `codex plugin list --json`, use the supported\ncachebuster\u002Freinstall flow, and start a new task. Hash exercised installed files\nagainst the source candidate. For Claude, identify the plugin source actually\nloaded by the CLI and verify its files the same way. A source\u002Fcache mismatch is\nan audit failure even if hooks otherwise work.\n\n## 2. Capture a pre-run baseline\n\nRecord:\n\n- `git status` and relevant file hashes;\n- existing session log, registry, handoff, lineage, and runtime markers;\n- scientific files and output state;\n- expected next log or session identifier when determinable.\n\nDo not delete suspicious state merely to make the test pass. If a clean-room\naudit is needed, create a fresh disposable repository.\n\n## 3. Launch one genuinely fresh host task\n\nRun the host CLI from the exact target root. Do not invoke any Mycelium hook\nscript in the task or immediately afterward. Use the canonical child-task\nprompt in `references\u002Fchild-task-prompt.md` — every probe it carries is\nrequired, including the nested-subagent probe, which may be marked N\u002FA only\nwhen the host exposes no native subagent mechanism. Ask the agent to expose\nthe exact automatic lifecycle context it receives and perform only the probes\nin the audit protocol. The launcher, not the child agent, owns artifact verification\nand baseline capture. Where the CLI supports it, disable skills in the child\ntask so the lifecycle-audit skill does not recursively launch or expand itself.\nKeep ordinary file-reading and editing tools available: the child needs them to\nobserve state and satisfy the expected Stop-compliance retry.\n\nExercise:\n\n1. SessionStart injection and active-state creation.\n2. When the host exposes a native subagent\u002Ftask mechanism, one read-only child\n   start and Stop while the primary remains active.\n3. A known-positive harmless analysis command.\n4. A near-neighbor negative command that must remain silent.\n5. One disposable successful edit using the host's normal edit tool. Remove it\n   with that tool when deletion is supported; otherwise, after edit-tracking\n   evidence is captured, use the protocol's single explicit shell cleanup.\n6. Natural Stop, including the expected enforcement block, compliant retry,\n   and a complete five-section handoff using Mycelium's exact headings once\n   each, in order, with nonblank section bodies.\n\nUse `python3` unless the target explicitly provides another interpreter; do not\nassume `python` exists. Preserve the real command exit status—an optional\ndependency failure may still be a valid lineage probe.\n\n## 4. Inspect only after natural Stop\n\nCompare post-run state with the baseline. Verify the first Stop blocked until\nthe disposable repository's living layer was updated, then verify exact-once finalization,\nregistry and lineage effects, handoff structure, cleanup, and absence of the\ndisposable file. Separate lifecycle-owned changes from agent-authored scientific\nchanges.\n\nClassify every unexpected result as one of:\n\n- **Artifact identity:** wrong source, version, cache, or packaged file.\n- **Host dispatch:** registration, trust, discovery, matcher, or host payload\n  never reached the hook.\n- **Hook logic:** dispatch occurred, but state or context violated the contract.\n- **Environment:** missing interpreter\u002Fdependency or CLI limitation unrelated to\n  Mycelium behavior.\n- **Agent compliance:** hook instructed the agent correctly, but it ignored or\n  rewrote required state.\n- **Inconclusive:** evidence was overwritten, manually stimulated, or never\n  captured.\n\nNever relabel an environment or compliance failure as a hook pass without\nshowing the hook evidence that did pass.\n\n## 5. Report and clean up\n\nReturn the evidence table from the audit protocol, exact errors, plugin identity,\nand a scientific-state diff. Remove only disposable probe files you created;\nretain failing lifecycle state when it is needed to reproduce a retry bug.\n\nDo not repair failures during the audit. Diagnose first so the observed state is\nnot contaminated. If a fix is requested afterward, switch to\n`$mycelium:develop`, add a red regression test, and repeat the black-box audit on\nthe new installed artifact.\n",{"name":5,"description":7},"main","2c9d7a461b75b93020c274bff2a508697dc05caa","skills\u002Flifecycle-audit",2,"https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory","https:\u002F\u002Frajlab.seas.upenn.edu\u002F","2026-11-15 02:57:15",{"source":8,"skillName":5,"commands":53,"prompt":63,"zipAvailable":64},[54,57,60],{"tool":55,"command":56},"npx","npx skills add arjunrajlaboratory\u002Fmycelium --skill lifecycle-audit",{"tool":58,"command":59},"bunx","bunx skills add arjunrajlaboratory\u002Fmycelium --skill lifecycle-audit",{"tool":61,"command":62},"pnpm","pnpm dlx skills add arjunrajlaboratory\u002Fmycelium --skill lifecycle-audit","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Flifecycle-audit\u002FSKILL.md",true,[66,72,76,80],{"path":67,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":70,"sourceHtmlUrl":71},"agents\u002Fopenai.yaml","REFERENCE","参考文档",235,"https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Flifecycle-audit\u002Fagents\u002Fopenai.yaml",{"path":73,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Faudit-protocol.md",8931,"https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Flifecycle-audit\u002Freferences\u002Faudit-protocol.md",{"path":77,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fchild-task-prompt.md",5154,"https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Flifecycle-audit\u002Freferences\u002Fchild-task-prompt.md",{"path":81,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":82,"sourceHtmlUrl":83},"resources\u002Fimplementation-playbook.md",16243,"https:\u002F\u002Fgithub.com\u002Fsapsapshen\u002Fai-memory-router\u002Fblob\u002Fmain\u002Fskills-store\u002Fcopilot\u002Fcode-documentation-doc-generate\u002Fresources\u002Fimplementation-playbook.md",[85,101,122,154,176,196],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":39},"01M6CVRE72Z220BXQCRXZT2481","transfer","Cross-pollinate learnings across Science subprojects by scanning recent .living\u002Flearnings.md entries and identifying actionable knowledge transfers. Use when user says \"transfer knowledge\", \"cross-pollinate\", \"sync learnings\", or \"knowledge transfer\".","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Ftransfer\u002FSKILL.md",[],[87,92,93,94,95,29,96,97,98,99,100],"sync","science","knowledge","user","scanning","knowledge transfer","learnings","across","entries",{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":39},"01M6CVRAK8SNRVYV5Q6FPMSK06","review","Review or audit analytical code and scientific changes in statistics, ML, bioinformatics, or data pipelines. Use for PRs, commits, diffs, sanity checks, correctness questions, pre-merge review, tripwire testing, or conversational grill mode. Consult the bundled statistical, leakage, bioinformatics, documentation, LLM-failure, and code-quality checklists; synthesize prioritized findings. Do not use for writing new analysis, reports, brainstorming, project initialization, or pure software review with no analytical component.","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Freview\u002FSKILL.md",[],[103,108,109,110,111,112,113,114,36,21,115,22,116,117,118,29,119,120,25,121],"testing","writing","diffs","component","documentation","brainstorming","ml","tripwire","commits","consult","code-quality","reports","grill","synthesize",{"skillNo":123,"name":124,"nameZh":6,"description":125,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":126,"categories":127,"keywords":138,"publishedAt":39},"01M6CVR761ZN49KWQXV3NG5AGA","report","Create a shareable report — a PDF, or a self-contained HTML report with a companion slide deck — a manuscript section, or a structured writeup from completed Mycelium analyses. Load the project's report templates, provenance, section guidance, and QC checklist. Use when the user wants a document artifact for collaborators or publication. Do not use for conversational explanation, standalone figures, performing analysis, ingestion, setup, or brainstorming.","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Freport\u002FSKILL.md",[128,133],{"categoryNo":129,"categoryName":130,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000002","文档处理","SCENARIO","使用场景",{"categoryNo":134,"categoryName":135,"taxonomyType":136,"taxonomyTypeDesc":137},"01JBSEEDCAT000000000000034","行政与办公人员","PROFESSION","适用职业",[124,139,140,113,141,142,143,144,22,145,29,146,147,148,149,150,25,151,152,153],"PDF","setup","create","templates","report templates","provenance","html report","deck","ingestion","guidance","checklist","document","artifact","self","mycelium",{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":163,"publishedAt":39},"01M6CVR0XNRWMMXX1N32741VE8","ingest","Add or import files and datasets into a Mycelium project with metadata, provenance, validation, and DATA_MANIFEST updates. Use for collaborator or facility outputs and formats such as CSV, FASTQ, XLSX, TIFF, JSON, FCS, TSV, or images. Do not use for analyzing existing data, listing datasets, deleting data, downloading URLs, or editing code.","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Fingest\u002FSKILL.md",[160],{"categoryNo":161,"categoryName":162,"taxonomyType":131,"taxonomyTypeDesc":132},"01JBSEEDCAT000000000000004","电子表格",[156,164,36,165,166,167,168,169,170,144,171,172,29,173,174,175,25,153],"xlsx","validation","metadata","csv","analyzing","data","manifest","json","editing","images","files","datasets",{"skillNo":177,"name":178,"nameZh":6,"description":179,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":180,"categories":181,"keywords":182,"publishedAt":39},"01M6CVQY7YVDS8PW82J1C67XJ3","ideas","Brainstorm research ideas using installed persona and ideation conventions. Use for fresh perspectives, creative angles, new analysis directions, expert viewpoints, \"what are we missing?\", or other open-ended exploration. Do not use for a defined analysis, code fix, project setup, or data ingestion task.","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Fideas\u002FSKILL.md",[],[178,183,140,184,169,22,185,186,187,188,189,29,147,190,191,192,193,25,194,195],"research","data ingestion","open","persona","brainstorm","perspectives","fix","ideation","creative","project setup","expert","exploration","conventions",{"skillNo":197,"name":198,"nameZh":6,"description":199,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":200,"categories":201,"keywords":202,"publishedAt":39},"01M6CVQTC883K58NHRMTS3DRF8","harden","Ship one learning's structural_mitigation_candidate as a real test, then mark the learning's mitigation_type structural. Picks a learning in .living\u002Flearnings.md that is still ambient-awareness but names a concrete check, writes the test in the repository's own test setup, proves it fails when the learning's problem is present and passes on the current code, and updates only that entry after the user confirms. Use when the user asks to harden, enforce, or turn a learning, gotcha, or lesson into a test or structural mitigation. Do not use for recording a new learning, promoting learnings to conventions, adding checks inside analysis code, or changing Mycelium itself.","https:\u002F\u002Fgithub.com\u002Farjunrajlaboratory\u002Fmycelium\u002Fblob\u002Fmain\u002Fskills\u002Fharden\u002FSKILL.md",[],[198,33,140,36,203,204,22,205,206,95,29,207,208,25,98,209,210,211,195,153],"repository","awareness","ship","learning","only","test setup","but","lesson","enforce"]