[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EBVWSS20R8T3W6PM2CK9NF":3,"skill-same:01M7EBVWSS20R8T3W6PM2CK9NF":61},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":16,"sourceHtmlUrl":17,"categories":18,"keywords":19,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M7EBVWSS20R8T3W6PM2CK9NF","review",null,"Review a branch, pull request or work in progress against a fixed point on two separate axes — Standards (does the diff follow the repository's written standards and the craft skills that apply to it?) and Spec (does it do what the originating issue or spec asked, no more and no less?). Use when reviewing a branch or PR, before opening a PR, or when asked to \"review since \u003Cref>\". Bugs are left to the harness's own reviewer.","conjurer-rich\u002Fskills","conjurer-rich","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F73941040?v=4","01M7EBNSGM9WBQSJWM8FQ5VNTR","Richard Allen","01M7EBNSH8C5824CFBXA3EK2H1",1,"MIT","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdelivery\u002Freview\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdelivery\u002Freview\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: review\ndescription: Review a branch, pull request or work in progress against a fixed point on two separate axes — Standards (does the diff follow the repository's written standards and the craft skills that apply to it?) and Spec (does it do what the originating issue or spec asked, no more and no less?). Use when reviewing a branch or PR, before opening a PR, or when asked to \"review since \u003Cref>\". Bugs are left to the harness's own reviewer.\n---\n\n# Review\n\nTwo-axis review of the diff between `HEAD` and a fixed point:\n\n- **Standards**: does the code follow this repository's written standards, and\n  the craft skills that govern the kind of code it is?\n- **Spec**: does the code do what the originating issue or spec asked?\n\nThe axes are reviewed separately and reported side by side, never merged, so a\npass on one cannot hide a failure on the other.\n\n**Not this skill:** finding correctness bugs. Use the harness's own reviewer\nfor that (`\u002Fcode-review` in Claude Code, `\u002Freview` or `codex review` in Codex)\nand run both when a change matters.\n\n## 1. Pin the fixed point\n\nUse the ref the user names (a commit, branch, tag, `main`, `HEAD~5`). With no\nref, use the merge-base with the default branch. A pull request number means\nits base branch.\n\nCapture the diff once: `git diff \u003Cfixed-point>...HEAD` (three dots, so the\ncomparison is against the merge-base), plus `git log \u003Cfixed-point>..HEAD\n--oneline`. Include uncommitted changes when the user asks to review work in\nprogress, and say that you did.\n\nConfirm the ref resolves (`git rev-parse \u003Cfixed-point>`) and the diff is not\nempty before going further. A bad ref or an empty diff stops the review here.\n\n## 2. Find the spec\n\nIn this order:\n\n1. Issue references in the commit messages or branch name (`#123`,\n   `Closes #45`), fetched from the repository's tracker.\n2. A path the user passed.\n3. A spec or plan file under `docs\u002F`, `specs\u002F`, `plans\u002F` or `.scratch\u002F` that\n   matches the branch or feature.\n4. Otherwise, ask where the spec is. If there is none, the Spec axis is\n   skipped and the report says \"no spec available\".\n\n## 3. Gather the standards\n\n- **The repository's written standards.** Every file that says how code\n  should be written: `CLAUDE.md` \u002F `AGENTS.md` at every level the diff\n  touches, `CODING_STANDARDS.md`, `CONTRIBUTING.md`, and docs those files\n  point to. When `CODING_STANDARDS.md` or `CONTRIBUTING.md` exists, it is on\n  the list.\n- **Up to three craft skills, chosen from the diff's traits.** For example:\n  `typescript-strict` for TypeScript, `testing` (and `front-end-testing` or\n  `react-testing`) when tests changed, `functional` for data transformations,\n  `hexagonal-architecture` or `domain-driven-design` only when the repository\n  has opted in, `api-design` for an externally consumed contract,\n  `structure-codebase` when files moved. Name the three you chose and why.\n- **The smell baseline** below, which applies even when the repository\n  documents nothing.\n\nTwo rules bind the baseline: **the repository overrides it** (where a written\nstandard endorses something the baseline would flag, drop the smell), and\n**each smell is a judgement call**, labelled as \"possible Feature Envy\", never\na hard violation. Skip anything tooling already enforces.\n\nFowler's code smells (*Refactoring*, chapter 3), each as *what it is* → *how to\nfix*:\n\n- **Mysterious Name**: a name that does not reveal what it does or holds →\n  rename it; if no honest name comes, the design is murky.\n- **Duplicated Code**: the same logic shape in more than one hunk or file →\n  extract the shared shape.\n- **Feature Envy**: a function reaching into another object's data more than\n  its own → move it to the data it envies.\n- **Data Clumps**: the same few fields or parameters travelling together →\n  give them one type.\n- **Primitive Obsession**: a primitive standing in for a domain concept → give\n  the concept its own small type.\n- **Repeated Switches**: the same `switch` or `if` cascade on the same type in\n  several places → one polymorphic call or one shared map.\n- **Shotgun Surgery**: one logical change forcing scattered edits → gather what\n  changes together.\n- **Divergent Change**: one module edited for several unrelated reasons →\n  split it.\n- **Speculative Generality**: abstraction or parameters for needs the spec\n  does not have → inline it until a real need shows.\n- **Message Chains**: long `a.b().c().d()` navigation → hide the walk behind\n  one method.\n- **Middle Man**: a module that mostly delegates → call the real target.\n- **Refused Bequest**: an implementer ignoring most of what it inherits →\n  composition instead of inheritance.\n\n## 4. Review both axes\n\nRun each axis in its own subagent, in parallel, so neither pollutes the other's\ncontext. Start each subagent with a fresh context window, never a fork of this\nsession: an axis that inherits the conversation inherits its opinions. Where the harness has no subagents, run Standards and then Spec, and\ndo not let the first colour the second. Each subagent receives the diff\ncommand and commit list.\n\n- **Standards** also receives the list of standard files, the chosen craft\n  skills (it loads them), and the smell baseline pasted in full. Brief:\n  \"Report, per file or hunk, (a) every place the diff breaks a written\n  standard or a loaded skill's rule, citing the file and rule; (b) any\n  baseline smell, named and quoted. Mark written-standard breaches as hard\n  violations where they are; smells are always judgement calls. Skip what\n  tooling enforces. Under 400 words.\"\n- **Spec** loads `acceptance-review` and runs it against the spec, with the\n  diff as the implementation under review. Brief: \"Report (a) criteria\n  missing or partly met, (b) behaviour in the diff nobody asked for, (c)\n  criteria that look met but whose implementation looks wrong. Quote the spec\n  for each. Under 400 words.\"\n\n## 5. Report\n\nPresent the two reports under `## Standards` and `## Spec`, verbatim or\nlightly cleaned. Do not merge or re-rank findings across axes.\n\nEnd with one line per axis: the number of findings and the worst one within\nthat axis. Do not pick a single winner across axes.\n","# Review\n\nTwo-axis review of the diff between `HEAD` and a fixed point:\n\n- **Standards**: does the code follow this repository's written standards, and\n  the craft skills that govern the kind of code it is?\n- **Spec**: does the code do what the originating issue or spec asked?\n\nThe axes are reviewed separately and reported side by side, never merged, so a\npass on one cannot hide a failure on the other.\n\n**Not this skill:** finding correctness bugs. Use the harness's own reviewer\nfor that (`\u002Fcode-review` in Claude Code, `\u002Freview` or `codex review` in Codex)\nand run both when a change matters.\n\n## 1. Pin the fixed point\n\nUse the ref the user names (a commit, branch, tag, `main`, `HEAD~5`). With no\nref, use the merge-base with the default branch. A pull request number means\nits base branch.\n\nCapture the diff once: `git diff \u003Cfixed-point>...HEAD` (three dots, so the\ncomparison is against the merge-base), plus `git log \u003Cfixed-point>..HEAD\n--oneline`. Include uncommitted changes when the user asks to review work in\nprogress, and say that you did.\n\nConfirm the ref resolves (`git rev-parse \u003Cfixed-point>`) and the diff is not\nempty before going further. A bad ref or an empty diff stops the review here.\n\n## 2. Find the spec\n\nIn this order:\n\n1. Issue references in the commit messages or branch name (`#123`,\n   `Closes #45`), fetched from the repository's tracker.\n2. A path the user passed.\n3. A spec or plan file under `docs\u002F`, `specs\u002F`, `plans\u002F` or `.scratch\u002F` that\n   matches the branch or feature.\n4. Otherwise, ask where the spec is. If there is none, the Spec axis is\n   skipped and the report says \"no spec available\".\n\n## 3. Gather the standards\n\n- **The repository's written standards.** Every file that says how code\n  should be written: `CLAUDE.md` \u002F `AGENTS.md` at every level the diff\n  touches, `CODING_STANDARDS.md`, `CONTRIBUTING.md`, and docs those files\n  point to. When `CODING_STANDARDS.md` or `CONTRIBUTING.md` exists, it is on\n  the list.\n- **Up to three craft skills, chosen from the diff's traits.** For example:\n  `typescript-strict` for TypeScript, `testing` (and `front-end-testing` or\n  `react-testing`) when tests changed, `functional` for data transformations,\n  `hexagonal-architecture` or `domain-driven-design` only when the repository\n  has opted in, `api-design` for an externally consumed contract,\n  `structure-codebase` when files moved. Name the three you chose and why.\n- **The smell baseline** below, which applies even when the repository\n  documents nothing.\n\nTwo rules bind the baseline: **the repository overrides it** (where a written\nstandard endorses something the baseline would flag, drop the smell), and\n**each smell is a judgement call**, labelled as \"possible Feature Envy\", never\na hard violation. Skip anything tooling already enforces.\n\nFowler's code smells (*Refactoring*, chapter 3), each as *what it is* → *how to\nfix*:\n\n- **Mysterious Name**: a name that does not reveal what it does or holds →\n  rename it; if no honest name comes, the design is murky.\n- **Duplicated Code**: the same logic shape in more than one hunk or file →\n  extract the shared shape.\n- **Feature Envy**: a function reaching into another object's data more than\n  its own → move it to the data it envies.\n- **Data Clumps**: the same few fields or parameters travelling together →\n  give them one type.\n- **Primitive Obsession**: a primitive standing in for a domain concept → give\n  the concept its own small type.\n- **Repeated Switches**: the same `switch` or `if` cascade on the same type in\n  several places → one polymorphic call or one shared map.\n- **Shotgun Surgery**: one logical change forcing scattered edits → gather what\n  changes together.\n- **Divergent Change**: one module edited for several unrelated reasons →\n  split it.\n- **Speculative Generality**: abstraction or parameters for needs the spec\n  does not have → inline it until a real need shows.\n- **Message Chains**: long `a.b().c().d()` navigation → hide the walk behind\n  one method.\n- **Middle Man**: a module that mostly delegates → call the real target.\n- **Refused Bequest**: an implementer ignoring most of what it inherits →\n  composition instead of inheritance.\n\n## 4. Review both axes\n\nRun each axis in its own subagent, in parallel, so neither pollutes the other's\ncontext. Start each subagent with a fresh context window, never a fork of this\nsession: an axis that inherits the conversation inherits its opinions. Where the harness has no subagents, run Standards and then Spec, and\ndo not let the first colour the second. Each subagent receives the diff\ncommand and commit list.\n\n- **Standards** also receives the list of standard files, the chosen craft\n  skills (it loads them), and the smell baseline pasted in full. Brief:\n  \"Report, per file or hunk, (a) every place the diff breaks a written\n  standard or a loaded skill's rule, citing the file and rule; (b) any\n  baseline smell, named and quoted. Mark written-standard breaches as hard\n  violations where they are; smells are always judgement calls. Skip what\n  tooling enforces. Under 400 words.\"\n- **Spec** loads `acceptance-review` and runs it against the spec, with the\n  diff as the implementation under review. Brief: \"Report (a) criteria\n  missing or partly met, (b) behaviour in the diff nobody asked for, (c)\n  criteria that look met but whose implementation looks wrong. Quote the spec\n  for each. Under 400 words.\"\n\n## 5. Report\n\nPresent the two reports under `## Standards` and `## Spec`, verbatim or\nlightly cleaned. Do not merge or re-rank findings across axes.\n\nEnd with one line per axis: the number of findings and the worst one within\nthat axis. Do not pick a single winner across axes.\n",{"name":5,"description":7},"main","9fac81929ac64b90ba023ba56f9f03fed440ebf4","skills\u002Fdelivery\u002Freview",0,"https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills","https:\u002F\u002Fgithub.com\u002Fconjurer-rich","https:\u002F\u002Fconjurersolutions.co.uk","2026-11-13 18:56:34",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add conjurer-rich\u002Fskills --skill review",{"tool":39,"command":40},"bunx","bunx skills add conjurer-rich\u002Fskills --skill review",{"tool":42,"command":43},"pnpm","pnpm dlx skills add conjurer-rich\u002Fskills --skill review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdelivery\u002Freview\u002FSKILL.md",true,[47,53,57],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"agents\u002Fopenai.yaml","REFERENCE","参考文档",185,"https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdelivery\u002Freview\u002Fagents\u002Fopenai.yaml",{"path":54,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":16},"LICENSE","许可证",1068,{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fsource-notes.md",1895,"https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fdelivery\u002Freview\u002Freferences\u002Fsource-notes.md",[62,70,78,86,94,102],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":66,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":20},"01M7EC238PW7ZVSEM3KT72ND70","wtf","Re-explain the immediately previous LLM response when it did not land, using plain, precise UK English.","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fwtf\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fwtf\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":74,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":20},"01M7EC20Q2R2QA6P90TXSGSKBG","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fwriting-for-agents\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fwriting-for-agents\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":82,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":20},"01M7EC1MRP6RBC3A15VHMB7JF0","technical-writing","Writing developer-facing prose that can be skimmed first and trusted enough to finish — READMEs, guides, tutorials, reference docs, proposals, PR descriptions, release notes. Load before drafting, rewriting, restructuring, or editing technical documents, when a doc reads as a wall of text, claims need receipts, or docs must serve AI agents. Covers reader-first structure, falsifiable claims, docs-as-behavior verification, and agent-readable reference shape. For sentence-level voice or generic AI-shaped prose use clarity; for plain English, lay readers, translation, or ASD-STE100 use simple-english; for recording a learning use expectations; for diagrams use diagrams; for API semantics use api-design.","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Ftechnical-writing\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Ftechnical-writing\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":90,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":20},"01M7EC1H1892N589M88AY2YN44","teach","Teach the user a new skill or concept, within this workspace.","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fteach\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Fteach\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":98,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":20},"01M7EC1BMMSYAE572Z40WP637S","find-skills","Discover and, with authorization, install agent skills from the open skills ecosystem. Use when the user explicitly asks to find or install a skill from the external skills ecosystem, asks whether an installable agent skill exists for a task, or wants to extend agent capabilities through skills.sh or a skill repository. Not for choosing among already-installed local skills or selecting software libraries, developer tools, applications, services, frameworks, or platform primitives; use evaluate-existing-solutions for technology choices.","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Ffind-skills\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fwriting\u002Ffind-skills\u002FSKILL.md",[],[],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":106,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":20},"01M7EBYV7CHGZZAN0NHGF9TWWH","reduce-system-complexity","Reduce total mechanism in a selected existing system or behavior path while conserving its agreed observable behavior and non-functional guarantees. Use when the user explicitly wants fewer branches, states, dependencies, layers, flags, retries, jobs, adapters, or operational moving parts and needs evidence that complexity was removed rather than relocated. Supports read-only diagnosis and authorized reduction slices. Not for routine cleanup, architecture-investment discovery, module-contract design, physical tree design, new behavior, functionality cuts, speculative rewrites, or Redux\u002Ffunctional reducer functions.","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Freduce-system-complexity\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fconjurer-rich\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Freduce-system-complexity\u002FSKILL.md",[],[]]