[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F93AW341G8H6C62CNN09JW":3,"skill-same:01M7F93AW341G8H6C62CNN09JW":81},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":29,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M7F93AW341G8H6C62CNN09JW","start-research",null,"Fan-out internet research into a verified, cited brief via persona research subagents plus an adversarial reviewer. Use to research a topic or gather prior art before a design.","szymonrychu\u002Fworkbench","szymonrychu","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F2943524?v=4","01M7F8VBGRDFJ2PSH14RM80E4G","Szymon Richert","01M7F8VBHCQTC2RMKXJ93RHMB3",0,"MIT","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: start-research\ndescription: \"Fan-out internet research into a verified, cited brief via persona research subagents plus an adversarial reviewer. Use to research a topic or gather prior art before a design.\"\n---\n\n# start-research\n\nResearches a topic across the internet using several persona research subagents\nin parallel and one adversarial reviewer subagent, then synthesizes a verified,\ncited brief at `docs\u002Fresearch\u002F\u003Cdate>-\u003Cslug>.md`. Spawns subagents via\nthe `Agent` tool - never the `Workflow` tool.\n\n## Procedure\n\n1. **Scope.** Restate the topic in one line. Pick the researchers from the\n   `lenses` argument (any subset of `docs-purist`, `practitioner`,\n   `prior-art-scout`, `paper-scout`):\n   - Default `lenses`: the three core personas (`docs-purist`, `practitioner`,\n     `prior-art-scout`). Add `paper-scout` when the topic has a research-paper\n     dimension.\n   - Broad topic: split one lens across two researchers with different\n     sub-scopes. A topic focus goes to `workbench:research-practitioner` as a\n     focus line in its prompt; no custom persona is run. Hard cap of 5\n     researchers.\n\n2. **Research (parallel).** Get each lens's output path first:\n   `bash ${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-research\u002Fscripts\u002Fdoc_path.sh research \"\u003Ctopic>-\u003Cpersona>\"`.\n   Spawn all researchers in a SINGLE message (multiple `Agent` calls) so they\n   run concurrently. Each is `subagent_type: workbench:research-\u003Cpersona>`\n   (e.g. `workbench:research-docs-purist`) with `model: sonnet` set explicitly\n   in the call. The persona's method and output contract live in its agent\n   file; the prompt carries only the topic and the output path for that\n   lens's brief.\n\n3. **Review (adversarial).** Get the reviewer's output path with\n   `doc_path.sh research \"\u003Ctopic>-review\"`. Spawn one\n   `subagent_type: workbench:research-reviewer` with `model: sonnet` set\n   explicitly. Its prompt carries only the topic, every lens brief path, and\n   its own output path. It marks each claim supported \u002F unsupported \u002F\n   contradicted and lists gaps and missing angles.\n\n4. **Follow-up (bounded).** If the reviewer reports material gaps, run ONE more\n   round of targeted researcher(s) scoped to those gaps, then a quick reviewer\n   re-check. Do not loop further.\n\n5. **Synthesize and write.** Drop reviewer-killed claims. Get the brief path:\n   ```\n   mise exec -- bash ${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-research\u002Fscripts\u002Fdoc_path.sh research \"\u003Ctopic>\"\n   ```\n   (`${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-research\u002Fscripts\u002Fbrief_path.sh \"\u003Ctopic>\"` is the equivalent path helper.) Write the\n   brief there and return its path plus an inline executive summary.\n\n## Brief format\n\nMarkdown: title + date + topic; executive summary; themed sections with inline\n`[n]` citations; explicit \"Best practices\" and \"Prior art\" sections; an \"Open\nquestions \u002F gaps\" section from the reviewer; a numbered \"Sources\" list. Every\nnon-trivial claim cites a source; killed claims do not appear.\n\nThe brief ends with a mandatory final `## Plan inputs` section, defined once in\nphase 1 of `${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-development\u002FSKILL.md` (lists: `reuse`,\n`do-not-build`, `fix-as-ac`, `ideas`).\n\n## Notes\n\n- Runnable standalone (`\u002Fstart-research \u003Ctopic>`) and called by\n  `start-development` during its discovery and design phases.\n- The four `research-*` skills (`research-docs`, `research-code-external`,\n  `research-code-internal`, `research-papers`) are the single-lens entry points;\n  use this skill when several lenses are wanted at once.\n- Always set each subagent's model explicitly (do not inherit the session\n  default).\n- `docs\u002Fresearch\u002F` is gitignored; briefs stay local.\n","# start-research\n\nResearches a topic across the internet using several persona research subagents\nin parallel and one adversarial reviewer subagent, then synthesizes a verified,\ncited brief at `docs\u002Fresearch\u002F\u003Cdate>-\u003Cslug>.md`. Spawns subagents via\nthe `Agent` tool - never the `Workflow` tool.\n\n## Procedure\n\n1. **Scope.** Restate the topic in one line. Pick the researchers from the\n   `lenses` argument (any subset of `docs-purist`, `practitioner`,\n   `prior-art-scout`, `paper-scout`):\n   - Default `lenses`: the three core personas (`docs-purist`, `practitioner`,\n     `prior-art-scout`). Add `paper-scout` when the topic has a research-paper\n     dimension.\n   - Broad topic: split one lens across two researchers with different\n     sub-scopes. A topic focus goes to `workbench:research-practitioner` as a\n     focus line in its prompt; no custom persona is run. Hard cap of 5\n     researchers.\n\n2. **Research (parallel).** Get each lens's output path first:\n   `bash ${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-research\u002Fscripts\u002Fdoc_path.sh research \"\u003Ctopic>-\u003Cpersona>\"`.\n   Spawn all researchers in a SINGLE message (multiple `Agent` calls) so they\n   run concurrently. Each is `subagent_type: workbench:research-\u003Cpersona>`\n   (e.g. `workbench:research-docs-purist`) with `model: sonnet` set explicitly\n   in the call. The persona's method and output contract live in its agent\n   file; the prompt carries only the topic and the output path for that\n   lens's brief.\n\n3. **Review (adversarial).** Get the reviewer's output path with\n   `doc_path.sh research \"\u003Ctopic>-review\"`. Spawn one\n   `subagent_type: workbench:research-reviewer` with `model: sonnet` set\n   explicitly. Its prompt carries only the topic, every lens brief path, and\n   its own output path. It marks each claim supported \u002F unsupported \u002F\n   contradicted and lists gaps and missing angles.\n\n4. **Follow-up (bounded).** If the reviewer reports material gaps, run ONE more\n   round of targeted researcher(s) scoped to those gaps, then a quick reviewer\n   re-check. Do not loop further.\n\n5. **Synthesize and write.** Drop reviewer-killed claims. Get the brief path:\n   ```\n   mise exec -- bash ${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-research\u002Fscripts\u002Fdoc_path.sh research \"\u003Ctopic>\"\n   ```\n   (`${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-research\u002Fscripts\u002Fbrief_path.sh \"\u003Ctopic>\"` is the equivalent path helper.) Write the\n   brief there and return its path plus an inline executive summary.\n\n## Brief format\n\nMarkdown: title + date + topic; executive summary; themed sections with inline\n`[n]` citations; explicit \"Best practices\" and \"Prior art\" sections; an \"Open\nquestions \u002F gaps\" section from the reviewer; a numbered \"Sources\" list. Every\nnon-trivial claim cites a source; killed claims do not appear.\n\nThe brief ends with a mandatory final `## Plan inputs` section, defined once in\nphase 1 of `${CLAUDE_PLUGIN_ROOT}\u002Fskills\u002Fstart-development\u002FSKILL.md` (lists: `reuse`,\n`do-not-build`, `fix-as-ac`, `ideas`).\n\n## Notes\n\n- Runnable standalone (`\u002Fstart-research \u003Ctopic>`) and called by\n  `start-development` during its discovery and design phases.\n- The four `research-*` skills (`research-docs`, `research-code-external`,\n  `research-code-internal`, `research-papers`) are the single-lens entry points;\n  use this skill when several lenses are wanted at once.\n- Always set each subagent's model explicitly (do not inherit the session\n  default).\n- `docs\u002Fresearch\u002F` is gitignored; briefs stay local.\n",{"name":5,"description":7},"main","ce8273d3492a5008dd408c4823db370b23e18863","plugins\u002Fworkbench\u002Fskills\u002Fstart-research","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench","https:\u002F\u002Fgithub.com\u002Fszymonrychu","www.szymonrichert.pl","2026-10-26 04:36:12",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add szymonrychu\u002Fworkbench --skill start-research",{"tool":37,"command":38},"bunx","bunx skills add szymonrychu\u002Fworkbench --skill start-research",{"tool":40,"command":41},"pnpm","pnpm dlx skills add szymonrychu\u002Fworkbench --skill start-research","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002FSKILL.md",true,[45,51,55,61,65,69,73,77],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"scripts\u002Fbrief_path.sh","SCRIPT","脚本",555,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Fscripts\u002Fbrief_path.sh",{"path":52,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":53,"sourceHtmlUrl":54},"scripts\u002Fdoc_path.sh",1087,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Fscripts\u002Fdoc_path.sh",{"path":56,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":59,"sourceHtmlUrl":60},"tests\u002Ftest_brief_path.bats","OTHER","其他",1935,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Ftests\u002Ftest_brief_path.bats",{"path":62,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":63,"sourceHtmlUrl":64},"tests\u002Ftest_brief_path.sh",2517,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Ftests\u002Ftest_brief_path.sh",{"path":66,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":67,"sourceHtmlUrl":68},"tests\u002Ftest_doc_path.bats",2590,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Ftests\u002Ftest_doc_path.bats",{"path":70,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":71,"sourceHtmlUrl":72},"tests\u002Ftest_doc_path.sh",3498,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Ftests\u002Ftest_doc_path.sh",{"path":74,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":75,"sourceHtmlUrl":76},"tests\u002Ftest_research_agents.py",2259,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Ftests\u002Ftest_research_agents.py",{"path":78,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":79,"sourceHtmlUrl":80},"tests\u002Ftest_start_skills_structure.py",18972,"https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fstart-research\u002Ftests\u002Ftest_start_skills_structure.py",[82,89,96,103,110,117],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M7F949PAG534XSY6BXH4GKY0","claudy","d claudy","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fharness\u002Fgolden\u002Fskills\u002Fclaudy\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M7F947QH8VQ5WGVMFWFS57M0","gamma","Literal line one.\nLiteral line two.","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fharness\u002Ffixtures\u002Ftree\u002Fplugins\u002Fextra\u002Fskills\u002Fgamma\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":19},"01M7F944VTWR1AAACZGTNWYJYP","beta","Quoted: description with a colon.","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fharness\u002Ffixtures\u002Ftree\u002Fplugins\u002Fcore\u002Fskills\u002Fbeta\u002FSKILL.md",[],[],{"skillNo":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":19},"01M7F941QWM4KQTX9KA6JQKF2G","alpha","Folded description that spans two lines and ends here.","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fscripts\u002Ftests\u002Fharness\u002Ffixtures\u002Ftree\u002Fplugins\u002Fcore\u002Fskills\u002Falpha\u002FSKILL.md",[],[],{"skillNo":111,"name":112,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":114,"categories":115,"keywords":116,"publishedAt":19},"01M7F93XXR4JY2S2KXD70JY4GM","worktree-merge","Rebases one subagent's worktree onto the base branch, optionally verifies the rebased tree, fast-forwards it into the worktree that has the base checked out (an epic integration tree or the main checkout), verifies every commit landed, and only then removes the worktree and its branch. Keeps the worktree intact on any conflict, dirty tree, or failed check. Use after a subagent returns work created with worktree-create.","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fworktree-merge\u002FSKILL.md",[],[],{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":19},"01M7F93TBQKKF3KM4VDVQ8PVHR","worktree-create","Creates one isolated git worktree for one subagent, at \u003Cworkbench>\u002F.claude\u002Fworktrees\u002F\u003Cname>, branched from main with the newest origin\u002Fmain and local main merged in. Refuses a name already in use so two agents can never share a tree. Use once per implementation or review subagent before dispatching it; pair with worktree-merge to land the work and clean up.","https:\u002F\u002Fgithub.com\u002Fszymonrychu\u002Fworkbench\u002Fblob\u002Fmain\u002Fplugins\u002Fworkbench\u002Fskills\u002Fworktree-create\u002FSKILL.md",[],[]]