[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M77NHD0GSKFAY8GW6RSN80TY":3,"skill-same:01M77NHD0GSKFAY8GW6RSN80TY":204},{"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":25,"commitSha":26,"skillDirectory":27,"forksCount":14,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M77NHD0GSKFAY8GW6RSN80TY","ce-compound",null,"Capture a solved problem's lesson in its cheapest durable form, a test or lint rule, a coding standard, or a repo learning, or refresh existing learnings against the codebase. Use when verified work corrected a wrong belief about how something behaves, such as a failed approach, a real defect a review caught, or a wrong assumption or document, or when auditing stale, overlapping, or drifted learnings. Do not auto-invoke it for work that went as first planned.","ucirello\u002Fcompound-engineering-plugin","ucirello","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F138827?v=4","01M77NFWYZ15SBS6EHVHBDDNA6","U Cirello","01M77NFWZJS5P3EKSVVKR8K5G9",0,"MIT","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: ce-compound\ndescription: Capture a solved problem's lesson in its cheapest durable form, a test or lint rule, a coding standard, or a repo learning, or refresh existing learnings against the codebase. Use when verified work corrected a wrong belief about how something behaves, such as a failed approach, a real defect a review caught, or a wrong assumption or document, or when auditing stale, overlapping, or drifted learnings. Do not auto-invoke it for work that went as first planned.\nargument-hint: \"[optional: brief context, or refresh and a scope] [mode:refresh] [mode:non-interactive] [depth:lightweight|full]\"\n---\n\n# \u002Fce-compound\n\n## Choosing the job\n\nThis skill does one of two jobs. **Capture** holds the lesson of one solved problem. **Refresh** audits the existing learnings against the current codebase and can delete docs and open a PR.\n\nRun refresh when the arguments contain `mode:refresh`, or when an interactive run was invoked to do one, as in `\u002Fce-compound refresh payments`, \"clean up the learnings\", or a standalone request to create `CONCEPTS.md`. An earlier remark in the session does not count. Read `references\u002Frefresh\u002Foverview.md` and follow it instead of the rest of this file. The refresh references name the parts of this file they reuse.\n\nWhen an interactive run is asked to create or update the project's coding standards or review docs as a whole, invoke the `ce-setup` skill with `standards` and stop.\n\nOtherwise run capture. A non-interactive run without `mode:refresh` is always capture, whatever its context says. Capture takes free-text context, so `refresh token rotation` after fixing that bug is a capture. If an interactive run's words could mean either job, ask once with the host's blocking question tool, and capture if there is no answer.\n\n## Capture\n\n**Outcome:** the lesson of one qualifying solved problem is held in the cheapest form that stops the repeat, grounded against the current tree, discoverable by the next agent.\n\n**Done:** the chosen form is written or updated and validated, vocabulary capture recorded, and the mode's completion report emitted; when no lesson qualifies, nothing is written and the report says why.\n\n**One learning per run.** A session that produced several gets several sequential runs, never one batched run. Read `references\u002Fresearch.md`; it explains what batching breaks.\n\n\n## Preconditions\n\nDocument only a problem that is solved and verified.\n\n\u003C!-- ce-durable-bar:start -->\nA lesson earns a durable form only when it holds durable project reasoning that is not readily recoverable from the current code, tests, checks, coding standards, comments, or documentation, and losing it would plausibly cause recurrence, material risk, or substantial rediscovery. Apply this counterfactual: without a record of this lesson, would a future engineer working from the current tree still be likely to repeat the mistake or redo substantial investigation? Completion, effort, and diff size do not establish eligibility.\n\u003C!-- ce-durable-bar:end -->\n\nIf the counterfactual fails, write nothing and report why. Judge this from the session rather than asking. An explicit invocation requests the judgment now but does not lower the bar.\n\nAn existing learning that became materially inaccurate or incomplete qualifies because leaving it would mislead. Update that learning instead of creating a duplicate.\n\n## Mode Detection\n\nEnter non-interactive mode when **either** holds: the arguments you were invoked with contain the `mode:non-interactive` token or its deprecated alias `mode:headless`, **or** the invocation makes non-interactive intent unmistakable, such as a caller or standing instruction asking to run `ce-compound` \"headless\", \"non-interactively\", \"unattended\", or \"without prompts\u002Fquestions\". Both tokens together is not a conflict. Bare \"automatically\" or \"auto-run\" is **not** on its own a non-interactive signal: it speaks to *invoking* the skill, not to suppressing its prompts. An ambiguous or absent signal defaults to interactive. Tokens starting with `mode:` or `depth:` are flags, not context: strip them before treating the remainder as the brief context hint. Once detected, non-interactive mode applies for the entire run.\n\n`references\u002Fmodes.md` decides depth, including what a `depth:` token does, as in `mode:non-interactive depth:lightweight` or `mode:non-interactive depth:full`. **Read it before step 1.**\n\n**Non-interactive mode asks nothing.** It asks no blocking question of any kind, in any phase, because a caller reaching this path has no human to answer one. Every non-interactive exit, including one taken before any phase runs, ends on a terminal signal a caller parses: `Documentation complete`, or `Documentation skipped` with the reason when the run wrote nothing. Interactive mode asks only where a step's own reference says to.\n\n## Artifact Root\n\nResolve `\u003Croot>` when you first compose a `\u003Croot>\u002Fsolutions\u002F` path, and pass a subagent the resolved path rather than the config.\n\n\u003C!-- ce-docs-root:start -->\n**Resolve the artifact root `\u003Croot>` before composing any artifact path.**\n\n- **Read** `docs_root` from `\u003Crepo-root>\u002F.rocketclaw\u002Fconfig.yaml` only (`\u003Crepo-root>` = `jj workspace root`). Do not read it from `config.local.yaml`. Unset -> `\u003Croot>` is `docs`, exactly as before.\n- **Validate** a set value: a repo-relative directory whose real, symlink-resolved path stays inside the repo and is neither the repo root nor under `.git\u002F`. Otherwise stop with an error naming `docs_root` and the value -- never fall back to `docs`.\n- **Use** `\u003Croot>` as the sole artifact location: create it if absent, compose each path as `\u003Croot>\u002F\u003Csubdir>` with this skill's own subdirectory, and never also read `docs`.\n\u003C!-- ce-docs-root:end -->\n\n## Write boundary\n\n**Only the orchestrator writes product files.** Phase 1 subagents write to per-run scratch only, and never touch `\u003Croot>\u002F`, project instruction files, or any other tracked path.\n\nThe orchestrator writes what `references\u002Fform.md` chooses, in every mode and without asking: at most one added test, added or tightened lint rule with any baseline entries it needs, or condition in an existing `CODING_STANDARDS.md`, and at most one learning under `\u003Croot>\u002Fsolutions\u002F`. It never changes product code. It also writes two maintenance side effects that their own step describes: `CONCEPTS.md` during vocabulary capture, and — **only in interactive Full mode after consent** — a small discoverability line in a project instruction file. Two further writes exist **only in interactive Full mode when the user selects them at the assembly destination step**: a rule file inside a writable declared Compound Pack, and the `packs:` entry appended to `.rocketclaw\u002Fconfig.yaml`. An instruction file is only ever edited, never created. Nothing else in the tree is written. Edits to *other* docs belong to a refresh, which a capture recommends or runs with a narrow scope.\n\n## Full Mode\n\n### Context-pressure handoff\n\nWhen an interactive run would otherwise choose Lightweight for context pressure, dispatch one fresh native OpenCode subagent loading `ce-compound` with `mode:non-interactive depth:full`. Read the source session ID with shell from `$OPENCODE_SESSION_ID`, never guess or recall it. Obtain the absolute workspace root with `jj workspace root`. Pass that ID, title when known, absolute root, any learning-doc path already written this run, and a one-sentence problem topic. Do not pre-digest the conversation. The delegated run is the orchestrator for the write boundary and reuses an already-written doc.\n\nThis skill explicitly requests delegation under a host rule allowing user- or skill-requested subagents; do not ask again under that rule. Respect unconditional prohibitions and actual denials. Use native `opencode.models` for configured model discovery when needed; model permission is separate from delegation permission, so omit an unauthorized model override and use a suitable inherited model. Classify actual errors separately as unavailable tool, permission denial, model-argument rejection, or exhausted nesting depth. Never evade restrictions through shell or another harness. Fall back to Lightweight only for actual unavailable\u002Fdenied\u002Fdepth delegation, naming which; advise re-running `ce-compound` from this same session once delegation is available, not in a fresh session. Explicit non-interactive `depth:` tokens remain authoritative.\n\nThe parent relays the delegated report and doc path, then offers with usual consent reported-but-unapplied interactive-only discoverability instruction-file edits and a Compound Pack destination. Do not run a second Full pass or re-edit the doc in the parent.\n\nRun these in order. Each reference is a required read at the step that names it.\n\n1. **Research** — read `references\u002Fresearch.md`.\n2. **Session history** — read `references\u002Fsession-history.md`, and start it *after* dispatching the Related Docs Finder so it overlaps your drafting rather than serializing.\n3. **Form, assembly, and write** — wait for every Phase 1 input, read `references\u002Fform.md` and write what it chooses, then read `references\u002Fassembly.md`.\n4. **Refresh check and discoverability** — read `references\u002Frefresh-and-discoverability.md`.\n5. **Optional enhancement** — read `references\u002Fenhancement.md`. Interactive only.\n6. **Report** — read `references\u002Freport.md` for the report shape your mode must emit and how the skill ends.\n\n**Lightweight Mode** replaces steps 1-6 with a single pass; read `references\u002Flightweight.md`, which defines its own completion output for both modes.\n","# \u002Fce-compound\n\n## Choosing the job\n\nThis skill does one of two jobs. **Capture** holds the lesson of one solved problem. **Refresh** audits the existing learnings against the current codebase and can delete docs and open a PR.\n\nRun refresh when the arguments contain `mode:refresh`, or when an interactive run was invoked to do one, as in `\u002Fce-compound refresh payments`, \"clean up the learnings\", or a standalone request to create `CONCEPTS.md`. An earlier remark in the session does not count. Read `references\u002Frefresh\u002Foverview.md` and follow it instead of the rest of this file. The refresh references name the parts of this file they reuse.\n\nWhen an interactive run is asked to create or update the project's coding standards or review docs as a whole, invoke the `ce-setup` skill with `standards` and stop.\n\nOtherwise run capture. A non-interactive run without `mode:refresh` is always capture, whatever its context says. Capture takes free-text context, so `refresh token rotation` after fixing that bug is a capture. If an interactive run's words could mean either job, ask once with the host's blocking question tool, and capture if there is no answer.\n\n## Capture\n\n**Outcome:** the lesson of one qualifying solved problem is held in the cheapest form that stops the repeat, grounded against the current tree, discoverable by the next agent.\n\n**Done:** the chosen form is written or updated and validated, vocabulary capture recorded, and the mode's completion report emitted; when no lesson qualifies, nothing is written and the report says why.\n\n**One learning per run.** A session that produced several gets several sequential runs, never one batched run. Read `references\u002Fresearch.md`; it explains what batching breaks.\n\n\n## Preconditions\n\nDocument only a problem that is solved and verified.\n\n\u003C!-- ce-durable-bar:start -->\nA lesson earns a durable form only when it holds durable project reasoning that is not readily recoverable from the current code, tests, checks, coding standards, comments, or documentation, and losing it would plausibly cause recurrence, material risk, or substantial rediscovery. Apply this counterfactual: without a record of this lesson, would a future engineer working from the current tree still be likely to repeat the mistake or redo substantial investigation? Completion, effort, and diff size do not establish eligibility.\n\u003C!-- ce-durable-bar:end -->\n\nIf the counterfactual fails, write nothing and report why. Judge this from the session rather than asking. An explicit invocation requests the judgment now but does not lower the bar.\n\nAn existing learning that became materially inaccurate or incomplete qualifies because leaving it would mislead. Update that learning instead of creating a duplicate.\n\n## Mode Detection\n\nEnter non-interactive mode when **either** holds: the arguments you were invoked with contain the `mode:non-interactive` token or its deprecated alias `mode:headless`, **or** the invocation makes non-interactive intent unmistakable, such as a caller or standing instruction asking to run `ce-compound` \"headless\", \"non-interactively\", \"unattended\", or \"without prompts\u002Fquestions\". Both tokens together is not a conflict. Bare \"automatically\" or \"auto-run\" is **not** on its own a non-interactive signal: it speaks to *invoking* the skill, not to suppressing its prompts. An ambiguous or absent signal defaults to interactive. Tokens starting with `mode:` or `depth:` are flags, not context: strip them before treating the remainder as the brief context hint. Once detected, non-interactive mode applies for the entire run.\n\n`references\u002Fmodes.md` decides depth, including what a `depth:` token does, as in `mode:non-interactive depth:lightweight` or `mode:non-interactive depth:full`. **Read it before step 1.**\n\n**Non-interactive mode asks nothing.** It asks no blocking question of any kind, in any phase, because a caller reaching this path has no human to answer one. Every non-interactive exit, including one taken before any phase runs, ends on a terminal signal a caller parses: `Documentation complete`, or `Documentation skipped` with the reason when the run wrote nothing. Interactive mode asks only where a step's own reference says to.\n\n## Artifact Root\n\nResolve `\u003Croot>` when you first compose a `\u003Croot>\u002Fsolutions\u002F` path, and pass a subagent the resolved path rather than the config.\n\n\u003C!-- ce-docs-root:start -->\n**Resolve the artifact root `\u003Croot>` before composing any artifact path.**\n\n- **Read** `docs_root` from `\u003Crepo-root>\u002F.rocketclaw\u002Fconfig.yaml` only (`\u003Crepo-root>` = `jj workspace root`). Do not read it from `config.local.yaml`. Unset -> `\u003Croot>` is `docs`, exactly as before.\n- **Validate** a set value: a repo-relative directory whose real, symlink-resolved path stays inside the repo and is neither the repo root nor under `.git\u002F`. Otherwise stop with an error naming `docs_root` and the value -- never fall back to `docs`.\n- **Use** `\u003Croot>` as the sole artifact location: create it if absent, compose each path as `\u003Croot>\u002F\u003Csubdir>` with this skill's own subdirectory, and never also read `docs`.\n\u003C!-- ce-docs-root:end -->\n\n## Write boundary\n\n**Only the orchestrator writes product files.** Phase 1 subagents write to per-run scratch only, and never touch `\u003Croot>\u002F`, project instruction files, or any other tracked path.\n\nThe orchestrator writes what `references\u002Fform.md` chooses, in every mode and without asking: at most one added test, added or tightened lint rule with any baseline entries it needs, or condition in an existing `CODING_STANDARDS.md`, and at most one learning under `\u003Croot>\u002Fsolutions\u002F`. It never changes product code. It also writes two maintenance side effects that their own step describes: `CONCEPTS.md` during vocabulary capture, and — **only in interactive Full mode after consent** — a small discoverability line in a project instruction file. Two further writes exist **only in interactive Full mode when the user selects them at the assembly destination step**: a rule file inside a writable declared Compound Pack, and the `packs:` entry appended to `.rocketclaw\u002Fconfig.yaml`. An instruction file is only ever edited, never created. Nothing else in the tree is written. Edits to *other* docs belong to a refresh, which a capture recommends or runs with a narrow scope.\n\n## Full Mode\n\n### Context-pressure handoff\n\nWhen an interactive run would otherwise choose Lightweight for context pressure, dispatch one fresh native OpenCode subagent loading `ce-compound` with `mode:non-interactive depth:full`. Read the source session ID with shell from `$OPENCODE_SESSION_ID`, never guess or recall it. Obtain the absolute workspace root with `jj workspace root`. Pass that ID, title when known, absolute root, any learning-doc path already written this run, and a one-sentence problem topic. Do not pre-digest the conversation. The delegated run is the orchestrator for the write boundary and reuses an already-written doc.\n\nThis skill explicitly requests delegation under a host rule allowing user- or skill-requested subagents; do not ask again under that rule. Respect unconditional prohibitions and actual denials. Use native `opencode.models` for configured model discovery when needed; model permission is separate from delegation permission, so omit an unauthorized model override and use a suitable inherited model. Classify actual errors separately as unavailable tool, permission denial, model-argument rejection, or exhausted nesting depth. Never evade restrictions through shell or another harness. Fall back to Lightweight only for actual unavailable\u002Fdenied\u002Fdepth delegation, naming which; advise re-running `ce-compound` from this same session once delegation is available, not in a fresh session. Explicit non-interactive `depth:` tokens remain authoritative.\n\nThe parent relays the delegated report and doc path, then offers with usual consent reported-but-unapplied interactive-only discoverability instruction-file edits and a Compound Pack destination. Do not run a second Full pass or re-edit the doc in the parent.\n\nRun these in order. Each reference is a required read at the step that names it.\n\n1. **Research** — read `references\u002Fresearch.md`.\n2. **Session history** — read `references\u002Fsession-history.md`, and start it *after* dispatching the Related Docs Finder so it overlaps your drafting rather than serializing.\n3. **Form, assembly, and write** — wait for every Phase 1 input, read `references\u002Fform.md` and write what it chooses, then read `references\u002Fassembly.md`.\n4. **Refresh check and discoverability** — read `references\u002Frefresh-and-discoverability.md`.\n5. **Optional enhancement** — read `references\u002Fenhancement.md`. Interactive only.\n6. **Report** — read `references\u002Freport.md` for the report shape your mode must emit and how the skill ends.\n\n**Lightweight Mode** replaces steps 1-6 with a single pass; read `references\u002Flightweight.md`, which defines its own completion output for both modes.\n",{"name":5,"description":7,"argument-hint":24},"[optional: brief context, or refresh and a scope] [mode:refresh] [mode:non-interactive] [depth:lightweight|full]","main","03ecf78d3885acd7cbd6ff9f37928d306cc7ba0e","skills\u002Fce-compound","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin","https:\u002F\u002Fgithub.com\u002Fucirello","Gopher. Mars aficionado.","2026-11-14 06:10:50",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add ucirello\u002Fcompound-engineering-plugin --skill ce-compound",{"tool":38,"command":39},"bunx","bunx skills add ucirello\u002Fcompound-engineering-plugin --skill ce-compound",{"tool":41,"command":42},"pnpm","pnpm dlx skills add ucirello\u002Fcompound-engineering-plugin --skill ce-compound","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002FSKILL.md",true,[46,52,58,62,66,70,74,78,82,86,90,94,98,102,106,110,114,118,122,126,130,134,138,142,146,150,154,158,162,166,170,174,180,184,188,192,196,200],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"assets\u002Fresolution-template.md","ASSET","资产文件",2842,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fassets\u002Fresolution-template.md",{"path":53,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":57},"references\u002Fagents\u002Fbest-practices-researcher.md","REFERENCE","参考文档",7302,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fagents\u002Fbest-practices-researcher.md",{"path":59,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":60,"sourceHtmlUrl":61},"references\u002Fagents\u002Fdata-integrity-guardian.md",3216,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fagents\u002Fdata-integrity-guardian.md",{"path":63,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":64,"sourceHtmlUrl":65},"references\u002Fagents\u002Fframework-docs-researcher.md",5551,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fagents\u002Fframework-docs-researcher.md",{"path":67,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":68,"sourceHtmlUrl":69},"references\u002Fagents\u002Fpattern-recognition-specialist.md",3675,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fagents\u002Fpattern-recognition-specialist.md",{"path":71,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":72,"sourceHtmlUrl":73},"references\u002Fagents\u002Fperformance-oracle.md",4588,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fagents\u002Fperformance-oracle.md",{"path":75,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":76,"sourceHtmlUrl":77},"references\u002Fagents\u002Fsecurity-sentinel.md",4159,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fagents\u002Fsecurity-sentinel.md",{"path":79,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Fagents\u002Fsession-historian.md",7312,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fagents\u002Fsession-historian.md",{"path":83,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Fassembly.md",16829,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fassembly.md",{"path":87,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":88,"sourceHtmlUrl":89},"references\u002Fconcepts-vocabulary.md",11688,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fconcepts-vocabulary.md",{"path":91,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":92,"sourceHtmlUrl":93},"references\u002Fenhancement.md",2738,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fenhancement.md",{"path":95,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":96,"sourceHtmlUrl":97},"references\u002Fform.md",4930,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fform.md",{"path":99,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":100,"sourceHtmlUrl":101},"references\u002Fgrounding-validation.md",7538,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fgrounding-validation.md",{"path":103,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":104,"sourceHtmlUrl":105},"references\u002Flightweight.md",7726,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Flightweight.md",{"path":107,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":108,"sourceHtmlUrl":109},"references\u002Fmodes.md",3211,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fmodes.md",{"path":111,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":112,"sourceHtmlUrl":113},"references\u002Frefresh-and-discoverability.md",9815,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh-and-discoverability.md",{"path":115,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":116,"sourceHtmlUrl":117},"references\u002Frefresh\u002Fclassify.md",9209,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Fclassify.md",{"path":119,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":120,"sourceHtmlUrl":121},"references\u002Frefresh\u002Fcommit.md",3496,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Fcommit.md",{"path":123,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":124,"sourceHtmlUrl":125},"references\u002Frefresh\u002Fdiscoverability.md",1877,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Fdiscoverability.md",{"path":127,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":128,"sourceHtmlUrl":129},"references\u002Frefresh\u002Finvestigate.md",4878,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Finvestigate.md",{"path":131,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":132,"sourceHtmlUrl":133},"references\u002Frefresh\u002Fmodes.md",3559,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Fmodes.md",{"path":135,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":136,"sourceHtmlUrl":137},"references\u002Frefresh\u002Foverview.md",5963,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Foverview.md",{"path":139,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":140,"sourceHtmlUrl":141},"references\u002Frefresh\u002Fper-action-flows.md",14909,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Fper-action-flows.md",{"path":143,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":144,"sourceHtmlUrl":145},"references\u002Frefresh\u002Freport.md",2069,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Freport.md",{"path":147,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":148,"sourceHtmlUrl":149},"references\u002Frefresh\u002Fscope.md",1935,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Fscope.md",{"path":151,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":152,"sourceHtmlUrl":153},"references\u002Frefresh\u002Fworth-audit.md",4695,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Frefresh\u002Fworth-audit.md",{"path":155,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":156,"sourceHtmlUrl":157},"references\u002Freport.md",8518,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Freport.md",{"path":159,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":160,"sourceHtmlUrl":161},"references\u002Fresearch.md",14484,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fresearch.md",{"path":163,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":164,"sourceHtmlUrl":165},"references\u002Fschema.yaml",9756,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fschema.yaml",{"path":167,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":168,"sourceHtmlUrl":169},"references\u002Fsession-history.md",10387,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fsession-history.md",{"path":171,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":172,"sourceHtmlUrl":173},"references\u002Fyaml-schema.md",9203,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Freferences\u002Fyaml-schema.md",{"path":175,"resourceType":176,"resourceTypeDesc":177,"sizeBytes":178,"sourceHtmlUrl":179},"scripts\u002Fpacks-resolve.py","SCRIPT","脚本",31580,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fscripts\u002Fpacks-resolve.py",{"path":181,"resourceType":176,"resourceTypeDesc":177,"sizeBytes":182,"sourceHtmlUrl":183},"scripts\u002Fsession-history\u002Fdiscover-sessions.sh",13095,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fscripts\u002Fsession-history\u002Fdiscover-sessions.sh",{"path":185,"resourceType":176,"resourceTypeDesc":177,"sizeBytes":186,"sourceHtmlUrl":187},"scripts\u002Fsession-history\u002Fextract-errors.py",10333,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fscripts\u002Fsession-history\u002Fextract-errors.py",{"path":189,"resourceType":176,"resourceTypeDesc":177,"sizeBytes":190,"sourceHtmlUrl":191},"scripts\u002Fsession-history\u002Fextract-metadata.py",24195,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fscripts\u002Fsession-history\u002Fextract-metadata.py",{"path":193,"resourceType":176,"resourceTypeDesc":177,"sizeBytes":194,"sourceHtmlUrl":195},"scripts\u002Fsession-history\u002Fextract-skeleton.py",23992,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fscripts\u002Fsession-history\u002Fextract-skeleton.py",{"path":197,"resourceType":176,"resourceTypeDesc":177,"sizeBytes":198,"sourceHtmlUrl":199},"scripts\u002Fvalidate-doc-claims.py",22133,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fscripts\u002Fvalidate-doc-claims.py",{"path":201,"resourceType":176,"resourceTypeDesc":177,"sizeBytes":202,"sourceHtmlUrl":203},"scripts\u002Fvalidate-frontmatter.py",5078,"https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-compound\u002Fscripts\u002Fvalidate-frontmatter.py",[205,212,219,226,233,240],{"skillNo":206,"name":207,"nameZh":6,"description":208,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":209,"categories":210,"keywords":211,"publishedAt":19},"01M78VM0YXEHCKHF8YT1P2WFHF","ce-retro","Look back over an agent session in this repo, the current one by default or any scope the user names across harnesses, and find what in the repo's environment keeps costing agents effort or leading them into mistakes, keeping only fixes the evidence shows are worth it. Lands the approved environment fixes and hands the rest to their owners. Use for a retro, postmortem, or papercut sweep of how agents worked in this repo. Use ce-compound to keep the lesson of one solved problem, and ce-debug for a bug in product code.","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Fskills\u002Fce-retro\u002FSKILL.md",[],[],{"skillNo":213,"name":214,"nameZh":6,"description":215,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":216,"categories":217,"keywords":218,"publishedAt":19},"01M77NMFCKBMKN3C8EACVGP8TG","release-notes","Draft release notes from merged pull requests since the last tag. Use when preparing a release.","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Fskill-eval-cell\u002Ffixtures\u002Freview-roster-skill-prose\u002Fskills\u002Frelease-notes\u002FSKILL.md",[],[],{"skillNo":220,"name":221,"nameZh":6,"description":222,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":223,"categories":224,"keywords":225,"publishedAt":19},"01M77NMEKMA0BXHC4230FV7S39","skill-one","Sample skill","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fskill-one\u002FSKILL.md",[],[],{"skillNo":227,"name":228,"nameZh":6,"description":229,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":230,"categories":231,"keywords":232,"publishedAt":19},"01M77NMCCVPH1F4P94W3CCAP5T","disabled-skill","A skill with model invocation disabled","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fdisabled-skill\u002FSKILL.md",[],[],{"skillNo":234,"name":235,"nameZh":6,"description":236,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":237,"categories":238,"keywords":239,"publishedAt":19},"01M77NMACM3ACB6PX92GMHFF3P","claude-only-skill","A skill restricted to Claude Code only","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fclaude-only-skill\u002FSKILL.md",[],[],{"skillNo":241,"name":242,"nameZh":6,"description":243,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":244,"categories":245,"keywords":246,"publishedAt":19},"01M77NM9ACPBEWHX6Z3QT1PRMS","agent-only-skill","An agent-only skill not intended for direct user invocation","https:\u002F\u002Fgithub.com\u002Fucirello\u002Fcompound-engineering-plugin\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsample-plugin\u002Fskills\u002Fagent-only-skill\u002FSKILL.md",[],[]]