[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79A3DPKCCDBQZ53G7M5SHK9":3,"skill-same:01M79A3DPKCCDBQZ53G7M5SHK9":47},{"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":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M79A3DPKCCDBQZ53G7M5SHK9","befund-stage-map",null,"Extracts the real import\u002Fuse graph per detected language, clusters files into stages by shallowest package boundary (never by manifest directory), and writes the full stage graph other befund skills depend on. Use when the user asks to \"map the repo's architecture\", \"show me the real module boundaries\", \"map stages and wires\", or as the first step of befund-autopilot's CHECK phase.","Anselmoo\u002Fwerkstoff","Anselmoo","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F13209783?v=4","01M799YTB3D7RZ4P3B8CSR2X85","Anselm Hahn","01M799YTBRAM6719D9V0NRGB4R",1,"MIT","https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Fplugins\u002Fbefund\u002Fskills\u002Fbefund-stage-map\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: befund-stage-map\ndescription: Extracts the real import\u002Fuse graph per detected language, clusters files into stages by shallowest package boundary (never by manifest directory), and writes the full stage graph other befund skills depend on. Use when the user asks to \"map the repo's architecture\", \"show me the real module boundaries\", \"map stages and wires\", or as the first step of befund-autopilot's CHECK phase.\n---\n\n# befund-stage-map\n\nBuild the repository's real architectural map: stages (package\u002Fmodule boundaries) and wires\n(the import\u002Fuse edges between them), verified against actual source, not guessed from\ndirectory names.\n\n## Step 0: Settings gate\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" check-enabled --repo \u003Crepo_root> --skill befund-stage-map\n```\n\nStop on non-zero exit. Carry the returned `output_dir` and `skip_verification` through every\nlater step.\n\n## Step 1: Detect languages\n\nReuse the same detection as preflight -- do not hand-roll a different threshold:\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" detect-languages --manifests \u003Cjson> --extension-counts \u003Cjson>\n```\n\n## Step 2: Extract the import graph, one inline command per language\n\nDispatch the `stage-mapper` agent once per detected language. Each dispatch runs the\nextraction as a **single inline read-only command** (e.g. a `grep -rn` \u002F `rg` pass over\nimport statements, or a language-native AST dump), never by creating scratch files -- that is\nthe agent's own refusal, restated here so the calling skill does not ask for anything that\nwould violate it.\n\nCluster files into stages by the **shallowest importable package boundary** in each language\n(the topmost directory that is itself an importable unit -- a directory with an `__init__.py`,\na `package.json`, a `go.mod`, etc.) -- never by \"which directory does the manifest that covers\nthis file live in,\" which is wrong whenever two packages share one manifest.\n\n## Step 3: Verify candidate wires\n\nUnless `skip_verification` is true, verify every candidate wire (an edge between two stages) by\nre-reading the actual import statement at its cited location before it is trusted -- do not\naccept a wire based on the extraction pass alone. Label unverified findings per\nskip-verification-behavior when `skip_verification` is true (no adversarial refutation, but\nevery finding still carries `verified: false`).\n\n## Step 4: Write the full graph, never sampled\n\nRule `stage-graph-vs-stage-map-json` requires `stage_graph.json` to carry the *complete* edge\ncount, not a sample -- `befund-arch-health` depends on accurate fan-in\u002Ffan-out. Build it\nwith top-level `stages`, `wires` (every edge), `edgeCount` (== `len(wires)`), and `deadEnds`\n(stages with no outgoing wires), then validate before writing:\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" validate-artifact --kind stage_graph --file \u003Cpath-or-inline-json>\n```\n\nThe validator rejects the artifact outright if `edgeCount != len(wires)` -- a sampled edge list\ncannot pass. The separate viewer-format `stage_map.json` (for STAGE_MAP.html) MAY sample down to\na handful of edges per stage for readability; only `stage_graph.json` carries this constraint.\n\n## Step 5: Write file_stage_index.json -- deliberately partial\n\nRule `file-stage-index-partial-coverage`: this flat `{file: stage}` lookup MUST contain only\nfiles that are BOTH an edge endpoint AND fall under a detected package boundary. A file that is\nneither is absent by design -- do not backfill it with a guessed stage. Downstream consumers\n(`befund-transform-brief`) treat a miss as `\"Unattributed\"`, never as an error; see\n`attribute-citation` in the CLI for that exact lookup behavior.\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" validate-artifact --kind file_stage_index --file \u003Cpath-or-inline-json>\n```\n\n## Step 6: Resolve output paths and write\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" resolve-output-path --repo \u003Crepo_root> --filename stage_graph.json\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" resolve-output-path --repo \u003Crepo_root> --filename file_stage_index.json\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" resolve-output-path --repo \u003Crepo_root> --filename stage_map_summary.json\n```\n\nWrite `STAGE_MAP.md`, `stage_graph.json`, `file_stage_index.json`, and `stage_map_summary.json`\nto their resolved paths. `stage_map_summary.json` is a reporting sidecar, not a load-bearing\nartifact: it records a short run summary -- stage\u002Fwire counts and a timestamp -- that\n`befund-status` reads only as a presence flag for its dashboard. No other befund skill\ndepends on its schema, so it is written directly here and never passed through\n`validate-artifact`.\n\n`stage_map.json` (viewer format, sampled edges) is not written by this skill at all: no step\nabove resolves a path for it or builds it, and `STAGE_MAP.html` below renders from the full\n`stage_graph.json` directly, never from `stage_map.json`'s sampled edges. `befund-transform-brief`\nis the sole producer of `stage_map.json` -- it creates the file, containing only a `flows` field,\nwhen it finds the file absent, and appends that same field without altering existing stage\u002Fwire\ndata when the file already exists.\n\nThen render `STAGE_MAP.html` -- a real canvas-based D3 viewer (force-directed graph layout,\npan\u002Fzoom, search, drag-to-reposition), not the old \"simple static graph render\":\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbuild_stage_map_html.py\" \\\n    --stage-graph \u003Cresolved stage_graph.json path> \\\n    --file-stage-index \u003Cresolved file_stage_index.json path> \\\n    --template ${CLAUDE_PLUGIN_ROOT}\u002Fassets\u002Fstage-map-viewer.html \\\n    --d3 ${CLAUDE_PLUGIN_ROOT}\u002Fassets\u002Finline-d3.html \\\n    --tokens ${CLAUDE_PLUGIN_ROOT}\u002Fassets\u002Ftokens.css \\\n    --out \u003Cresolved STAGE_MAP.html path>\n```\n\nThis reads the FULL `stage_graph.json` directly (never `stage_map.json`'s sampled edges --\ncanvas rendering with proper culling doesn't need the edge count thinned down for readability\nthe way the old prose-only render did), and reuses `lib.graph`'s own `find_cycles`\u002F\n`find_god_modules` to highlight cycles and god-modules in the viewer -- the exact same\nfunctions `befund-arch-health` reports from, so the viewer can never show something that\nskill disagrees with.\n\n## Read-only constraint\n\nNever use Write\u002FEdit for anything outside the resolved output paths above. The extraction\ncommand dispatched to `stage-mapper` is read-only (grep\u002FAST-dump), never a scratch file or a\nmutation.\n","# befund-stage-map\n\nBuild the repository's real architectural map: stages (package\u002Fmodule boundaries) and wires\n(the import\u002Fuse edges between them), verified against actual source, not guessed from\ndirectory names.\n\n## Step 0: Settings gate\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" check-enabled --repo \u003Crepo_root> --skill befund-stage-map\n```\n\nStop on non-zero exit. Carry the returned `output_dir` and `skip_verification` through every\nlater step.\n\n## Step 1: Detect languages\n\nReuse the same detection as preflight -- do not hand-roll a different threshold:\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" detect-languages --manifests \u003Cjson> --extension-counts \u003Cjson>\n```\n\n## Step 2: Extract the import graph, one inline command per language\n\nDispatch the `stage-mapper` agent once per detected language. Each dispatch runs the\nextraction as a **single inline read-only command** (e.g. a `grep -rn` \u002F `rg` pass over\nimport statements, or a language-native AST dump), never by creating scratch files -- that is\nthe agent's own refusal, restated here so the calling skill does not ask for anything that\nwould violate it.\n\nCluster files into stages by the **shallowest importable package boundary** in each language\n(the topmost directory that is itself an importable unit -- a directory with an `__init__.py`,\na `package.json`, a `go.mod`, etc.) -- never by \"which directory does the manifest that covers\nthis file live in,\" which is wrong whenever two packages share one manifest.\n\n## Step 3: Verify candidate wires\n\nUnless `skip_verification` is true, verify every candidate wire (an edge between two stages) by\nre-reading the actual import statement at its cited location before it is trusted -- do not\naccept a wire based on the extraction pass alone. Label unverified findings per\nskip-verification-behavior when `skip_verification` is true (no adversarial refutation, but\nevery finding still carries `verified: false`).\n\n## Step 4: Write the full graph, never sampled\n\nRule `stage-graph-vs-stage-map-json` requires `stage_graph.json` to carry the *complete* edge\ncount, not a sample -- `befund-arch-health` depends on accurate fan-in\u002Ffan-out. Build it\nwith top-level `stages`, `wires` (every edge), `edgeCount` (== `len(wires)`), and `deadEnds`\n(stages with no outgoing wires), then validate before writing:\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" validate-artifact --kind stage_graph --file \u003Cpath-or-inline-json>\n```\n\nThe validator rejects the artifact outright if `edgeCount != len(wires)` -- a sampled edge list\ncannot pass. The separate viewer-format `stage_map.json` (for STAGE_MAP.html) MAY sample down to\na handful of edges per stage for readability; only `stage_graph.json` carries this constraint.\n\n## Step 5: Write file_stage_index.json -- deliberately partial\n\nRule `file-stage-index-partial-coverage`: this flat `{file: stage}` lookup MUST contain only\nfiles that are BOTH an edge endpoint AND fall under a detected package boundary. A file that is\nneither is absent by design -- do not backfill it with a guessed stage. Downstream consumers\n(`befund-transform-brief`) treat a miss as `\"Unattributed\"`, never as an error; see\n`attribute-citation` in the CLI for that exact lookup behavior.\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" validate-artifact --kind file_stage_index --file \u003Cpath-or-inline-json>\n```\n\n## Step 6: Resolve output paths and write\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" resolve-output-path --repo \u003Crepo_root> --filename stage_graph.json\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" resolve-output-path --repo \u003Crepo_root> --filename file_stage_index.json\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbefund_cli.py\" resolve-output-path --repo \u003Crepo_root> --filename stage_map_summary.json\n```\n\nWrite `STAGE_MAP.md`, `stage_graph.json`, `file_stage_index.json`, and `stage_map_summary.json`\nto their resolved paths. `stage_map_summary.json` is a reporting sidecar, not a load-bearing\nartifact: it records a short run summary -- stage\u002Fwire counts and a timestamp -- that\n`befund-status` reads only as a presence flag for its dashboard. No other befund skill\ndepends on its schema, so it is written directly here and never passed through\n`validate-artifact`.\n\n`stage_map.json` (viewer format, sampled edges) is not written by this skill at all: no step\nabove resolves a path for it or builds it, and `STAGE_MAP.html` below renders from the full\n`stage_graph.json` directly, never from `stage_map.json`'s sampled edges. `befund-transform-brief`\nis the sole producer of `stage_map.json` -- it creates the file, containing only a `flows` field,\nwhen it finds the file absent, and appends that same field without altering existing stage\u002Fwire\ndata when the file already exists.\n\nThen render `STAGE_MAP.html` -- a real canvas-based D3 viewer (force-directed graph layout,\npan\u002Fzoom, search, drag-to-reposition), not the old \"simple static graph render\":\n\n```\npython3 -B \"${CLAUDE_PLUGIN_ROOT}\u002Fscripts\u002Fbuild_stage_map_html.py\" \\\n    --stage-graph \u003Cresolved stage_graph.json path> \\\n    --file-stage-index \u003Cresolved file_stage_index.json path> \\\n    --template ${CLAUDE_PLUGIN_ROOT}\u002Fassets\u002Fstage-map-viewer.html \\\n    --d3 ${CLAUDE_PLUGIN_ROOT}\u002Fassets\u002Finline-d3.html \\\n    --tokens ${CLAUDE_PLUGIN_ROOT}\u002Fassets\u002Ftokens.css \\\n    --out \u003Cresolved STAGE_MAP.html path>\n```\n\nThis reads the FULL `stage_graph.json` directly (never `stage_map.json`'s sampled edges --\ncanvas rendering with proper culling doesn't need the edge count thinned down for readability\nthe way the old prose-only render did), and reuses `lib.graph`'s own `find_cycles`\u002F\n`find_god_modules` to highlight cycles and god-modules in the viewer -- the exact same\nfunctions `befund-arch-health` reports from, so the viewer can never show something that\nskill disagrees with.\n\n## Read-only constraint\n\nNever use Write\u002FEdit for anything outside the resolved output paths above. The extraction\ncommand dispatched to `stage-mapper` is read-only (grep\u002FAST-dump), never a scratch file or a\nmutation.\n",{"name":5,"description":7},"main","6158c1e4e112c06af2c7ac01b13b2ed7214c0fcf","plugins\u002Fbefund\u002Fskills\u002Fbefund-stage-map",0,"https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff","https:\u002F\u002Fanselmoo.github.io\u002Fwerkstoff\u002F","https:\u002F\u002Fgithub.com\u002FAnselmoo","Likes to Develop","2026-11-02 04:20:59",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add Anselmoo\u002Fwerkstoff --skill befund-stage-map",{"tool":39,"command":40},"bunx","bunx skills add Anselmoo\u002Fwerkstoff --skill befund-stage-map",{"tool":42,"command":43},"pnpm","pnpm dlx skills add Anselmoo\u002Fwerkstoff --skill befund-stage-map","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Fplugins\u002Fbefund\u002Fskills\u002Fbefund-stage-map\u002FSKILL.md",true,[],[48,55,62,67,72,77],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M79AJTRBNBX03WWKYJ8ENM45","toy-check","A skill that checks things in the repository and reports what it finds.","https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Ftest\u002Fworkflows\u002Ffixtures\u002Ftoy-plugin\u002Fskills\u002Ftoy-check\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M79AJS292F3HTDRXWZVRSN5B","demo-check","Reports which generated files changed since the last generation run. Use when the user asks what drifted. Read-only; never edits a file.","https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Ftest\u002Fplugins\u002Ffixtures\u002Fnacharbeit-uncompleted-run\u002Fplugins\u002Fdemo\u002Fskills\u002Fdemo-check\u002FSKILL.md",[],[],{"skillNo":63,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M79AJP4TFQXKT1AAK1WC1JYK","https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Ftest\u002Fplugins\u002Ffixtures\u002Fnacharbeit-stale-lock\u002Fplugins\u002Fdemo\u002Fskills\u002Fdemo-check\u002FSKILL.md",[],[],{"skillNo":68,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M79AJM6Y50WP3G2BR4H0M60Z","https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Ftest\u002Fplugins\u002Ffixtures\u002Fnacharbeit-preflight-mixed\u002Fplugins\u002Fdemo\u002Fskills\u002Fdemo-check\u002FSKILL.md",[],[],{"skillNo":73,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M79AJGR6DMT8KCCK071V78XM","https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Ftest\u002Fplugins\u002Ffixtures\u002Fnacharbeit-hooks-shape\u002Fplugins\u002Fdemo\u002Fskills\u002Fdemo-check\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M79AJBJ4CHC4PZVWKERWF3HF","zirkel-verify-assumptions","Checks exactly ONE named assumption against real evidence in at most 3 steps and reports one outcome: confidence raised (>=90), reading changed, or still unresolved. Use to resolve a single load-bearing uncertainty — typically a blocking entry surfaced by zirkel-clarify-scope: \"verify this one assumption\", \"is it actually true that X\", \"check this before we rely on it\". For multiple uncertainties, invoke once per uncertainty — the step budget is never shared.","https:\u002F\u002Fgithub.com\u002FAnselmoo\u002Fwerkstoff\u002Fblob\u002Fmain\u002Fplugins\u002Fzirkel\u002Fskills\u002Fzirkel-verify-assumptions\u002FSKILL.md",[],[]]