[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M73GQ07SAXG2WVEMKJW5FDAT":3,"skill-same:01M73GQ07SAXG2WVEMKJW5FDAT":69},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":47,"commitSha":48,"skillDirectory":49,"forksCount":50,"repositoryHtmlUrl":51,"repositoryHomepageUrl":6,"ownerHtmlUrl":52,"ownerBlogUrl":6,"ownerDescription":53,"updatedAt":54,"install":55,"resources":68},"01M73GQ07SAXG2WVEMKJW5FDAT","skill-graph-analytics",null,"Weekly fleet analytics over the per-project skill-graph projections: consolidate root + Projects\u002F\u003Cco-x> graphs into a dated snapshot, triage skill convergence (consolidation\u002Fpromotion candidates), delivery drift (root skills missing from projects), and root-graph orphans (4-way cross-check), and record findings. Use when: the weekly analytics cadence fires (manual until the Friday full-override revival — see When to Run), the user says \"skill-graph analytics\", \"fleet skill graph\", \"skill graph report\", or asks which skills the project fleet carries, which root skills never reached projects, whether any agent or skill is orphaned, or what changed fleet-wide since last week.","5throck\u002Fai-workspace-standards","5throck","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F10958964?v=4","01M73BFEH0V0VS483TW4JD49JW","01M73BFEJ5VQZJHJSFTVPAZGDM",25,"AGPL-3.0","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fcommon\u002F.claude\u002Fskills\u002Fskill-graph-analytics\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: skill-graph-analytics\nversion: 1.4.0\ndescription: >\n  Weekly fleet analytics over the per-project skill-graph projections:\n  consolidate root + Projects\u002F\u003Cco-x> graphs into a dated snapshot, triage\n  skill convergence (consolidation\u002Fpromotion candidates), delivery drift\n  (root skills missing from projects), and root-graph orphans (4-way\n  cross-check), and record findings. Use when: the weekly analytics cadence\n  fires (manual until the Friday full-override revival — see When to Run), the user says \"skill-graph\n  analytics\", \"fleet skill graph\", \"skill graph report\", or asks which skills\n  the project fleet carries, which root skills never reached projects, whether\n  any agent or skill is orphaned, or what changed fleet-wide since last week.\nstatus: active\nscope: workspace\nl2_propagate: false\nowner: pm\nlast_reviewed: 2026-11-08\nprerequisites: workspace-root CWD; docs\u002Fskill-graph.json + Projects\u002F\u003Cco-x>\u002Fdocs\u002Fskill-graph.json projections present\nrelates_to:\n  - skill: project-resync\n    type: composes_with\n  - skill: context-commonization-review\n    type: relates_to\nmetadata:\n  type: process\n  triggers:\n    - skill-graph analytics\n    - fleet skill graph\n    - skill graph report\n    - skill graph fleet report\n    - skill convergence triage\n    - orphan agent check\n    - orphan skill check\n---\n\n# Skill: skill-graph-analytics\n\nWeekly fleet analytics over the per-project skill-graph projections\n(`Projects\u002F\u003Cco-x>\u002Fdocs\u002Fskill-graph.json`, root `docs\u002Fskill-graph.json`).\nOne run answers: which skills does the fleet carry, which root skills never\nreached the projects, how far has each project drifted from the root\nprojection, and what is NEW or VANISHED fleet-wide since the last snapshot?\n\n**Triage only — never auto-modify.** This skill produces tickets and findings;\nit does not edit skills, promote content, or rewrite graphs.\n\n## When to Run\n\n- **Weekly cadence** — NO scheduled runner invokes this skill as of 2026-11-03\n  (automation roster: 01:30 fleet-review+resync, 04:00 governance-ticket —\n  neither carries the weekly consolidation; the old 02:30 reference here rotted,\n  T-20261101-004). **Revival home DECIDED 2026-11-03 (T-20261101-004)**: the\n  01:30 fleet runner's FRIDAY FULL OVERRIDE carries the weekly consolidation\n  (one prompt-line addition, no new schedule; runner prompt updated via\n  automation tooling). Until that override next fires, run it manually at the\n  start of a weekly maintenance session.\n- **On demand** — after a fleet-wide skill promotion\u002Fupgrade wave, before a\n  consolidation decision, or whenever skill-graph drift is suspected.\n\n## Step-by-Step Procedure\n\n### 1. Run the fleet report\n\n```bash\nbun scripts\u002Fskill-graph-fleet-report.ts\n```\n\nRead-only over all inputs. Writes one machine snapshot per local calendar date:\n`memory\u002Fskill-graph-metrics\u002Fsnapshot-\u003CYYYY-MM-DD>.json` (same-date reruns\noverwrite). First run establishes the baseline — no NEW\u002FVANISHED diff exists yet.\nProjects without a graph projection are skipped with a note; zero graphs found\nexits 1 (fix the generator, `bun scripts\u002Fgenerate-skill-graph.ts`, before triage).\n**Fleet membership SSOT** (DEC-20261107-01): the report's project set comes from\n`lib\u002Ffleet.ts` — `Projects\u002Fco-*` directories minus the exclusions recorded in\n`docs\u002Ffleet-membership.json` (verify with `bun scripts\u002Ffleet-list.ts --json`).\nNever derive the set with a raw `Projects\u002Fco-*` glob; convergence denominators\ncount members only.\n\n### 2. Read the snapshot and the report\n\n- **Per-project stats + Jaccard distance vs root** — bigger distance = narrower\n  or more divergent projection; small projects are expected to diverge, a large\n  project with rising distance is a delivery-gap signal.\n- **Presence matrix + top skills** — fleet presence counts per skill\n  **capability** (frontmatter `capability`, default = skill name; v1 project\n  graphs derive it from the bare node id, so v1 and v2 graphs join on the\n  same key).\n- **Skill convergence by capability** — capabilities carried by >= 3 project\n  graphs, split into `converged` (one content hash), `convergence-with-\n  divergence` (>= 2 hashes) and `hash-unknown` (a v1 project graph carries no\n  hashes yet); see §3 and §3.4.\n- **Root skills missing per project** — the delivery-drift surface.\n- **Root-graph orphan candidates** — graph-isolated root skills\u002Fagents with the\n  4-way cross-check axes (see §3.5); feeds the orphan triage rules.\n- **NEW\u002FVANISHED vs the previous snapshot** — fleet-wide skill evolution.\n\n### 3. Triage (judgment — not automatable)\n\n| Signal | Threshold | Action |\n|---|---|---|\n| **Capability** present in project graphs but **absent from the root graph**, `converged` class (single content hash) | >= 3 project graphs | Consolidation\u002Fpromotion candidate — file a ticket: `bun scripts\u002Fticket.ts create --manual \"skill-graph: \u003Ccapability> converged in N projects, absent from root — consolidation review\" --priority normal` (routes through §3.7.5 governance-backlog triage) |\n| Capability in the `convergence-with-divergence` class (>= 3 projects, >= 2 content hashes) | any | **Reconciliation candidate, not a promotion candidate** — see §3.4; never auto-promote |\n| Capability in the `hash-unknown` class | >= 3 project graphs | Treat as a promotion *candidate* by presence only; content equality is unverified until the carrying v1 projects are upgraded (`upgrade-project` delivers the v2 generator) — compare the SKILL.md files by hand before filing a promotion ticket |\n| **Root skill missing from many projects** | missing from >= 3 project graphs | Delivery-gap candidate — file a ticket naming the skill and the missing projects (likely template-delivery drift; check `templates\u002Fcommon\u002Fskills\u002F` and the variant contract) |\n| NEW\u002FVANISHED skills in the diff | any | Verify each is intentional (new skill landed \u002F skill deprecated) — unexplained entries are tickets |\n| **Root-graph orphan candidate** (report section \"Root-graph orphan candidates\") | any isolated skill\u002Fagent | Apply the 4-way orphan criteria below — never treat graph isolation alone as proof of orphanhood |\n| Everything else | — | No action; the snapshot is the record |\n\n### 3.4 Divergence triage (capability-based convergence)\n\nSince skill-graph v2 (ADR-0060 Amendment 11) a skill node's identity is\n`skill:\u003Cscope>\u002F\u003Cname>` with a `content_hash`, and fleet convergence groups by\n**capability** rather than by raw node id, so same-named skills whose content\ndiffers no longer look identical. For each `convergence-with-divergence`\ncapability in the report:\n\n1. Read the hash groups (`hash: projects`) and diff the SKILL.md copies of the\n   largest and smallest group.\n2. Classify: **fork** (intentional variant specialisation — set a distinct\n   `capability:` in the forked SKILL.md frontmatter, or keep and record why),\n   **drift** (accidental divergence — reconcile to one copy, then the group\n   becomes `converged` and a promotion candidate), or **rename** (same job,\n   different name — set the same `capability:` on both so they group).\n3. File one reconciliation ticket per capability (`bun scripts\u002Fticket.ts create\n   --manual \"skill-graph: \u003Ccapability> diverged across N projects — reconcile\"\n   --priority normal`); never edit the project copies from this skill.\n\nThe root `docs\u002Fskill-graph.md` section \"Same-Name Divergence (E2)\" and\n`bun scripts\u002Fverify-skill-graph.ts` carry the same signal inside the workspace\n(`version-drift` = same version, different content — bump or reconcile).\n`bun scripts\u002Fgenerate-skill-graph.ts --impact \u003Cskill>` lists the agents,\nprocedures, scopes, tests and contracts a change to a skill touches.\n\n### 3.5 Root-graph orphan cross-check (4-way criteria)\n\nA root skill or agent is an **orphan candidate** only when it is graph-isolated\n(zero edges — reported in the fleet report's orphan section since\nfleet-report v1.1.0). Isolation alone is NOT proof: classify each candidate by\ncrossing four axes, all reported in the same section:\n\n| Axis | Meaning | How the report shows it |\n|------|---------|------------------------|\n| **Definition** | The node exists in the graph (skills\u002F, agents\u002F) | the isolation list itself |\n| **Registry** | A SKILLS.md \u002F roster row exists | `registry ✓\u002F✗` |\n| **Mirror** | The 4 platform skill bases carry it | `mirrors N\u002F4` |\n| **Reference** | Workflow docs mention it (bounded corpus: platform\u002Fagent context docs + procedures\u002F + process\u002F) | `refs N` |\n\nTriage rules:\n\n- **All four axes present** (registry ✓, mirrors 4\u002F4, refs > 0) → *graph-edge\n  gap*, not an orphan. Fix the graph, not the skill: add a hand-reasoned\n  `docs\u002Fskill-graph.overrides.json` entry (`relates_to`, with `reason`\u002F`since`)\n  or a workflow-doc citation, then regenerate.\n- **Definition + Registry only** (no mirror, no refs) → *weak candidate* —\n  wire it into a workflow or schedule deprecation review.\n- **Missing registry row or mirrors** → *delivery drift* — the standard\n  registry\u002Fmirror fixes apply before any orphan verdict.\n- **Zero axes beyond definition** → *true orphan* — retire via the\n  skill\u002Fagent lifecycle manager.\n- Agent candidates: mirror axis does not apply (agents are not mirrored);\n  weigh roster presence (AGENTS.md §1\u002F§4.1) and doc references instead.\n\nHistorical baseline (2026-09-23 audit): 12 L0 + 48 variant skills were\ngraph-isolated; the workflow-doc citation scan (generator Source 4.8) plus four\noverrides entries brought L0 isolation to zero — variant-isolated skills remain\non this cadence's triage list.\n\n**Centrality-readiness note (Phase 2, future work)**: presence counts are a\nblunt proxy for importance. A later phase adds centrality ranking over the root\ngraph (degree\u002Fbetweenness over `step_uses_skill`\u002F`used_by` edges) per\ndocs\u002Fdesigns\u002F2026-09-22-skill-graph-fleet-analytics-design.md — do not\nover-prioritize a skill on presence alone today.\n\n### 4. Record findings\n\nAppend a `## Skill-Graph Fleet Analytics` section to the daily memory log\n(`memory\u002FYYYY-MM-DD.md`): snapshot path, headline stats (projects covered,\nfleet skill count, top-3 by presence), every ticket filed, and any\ndelivery-gap or convergence observation that did not merit a ticket.\n\n## Notes\n\n- Capability-keyed presence: presence is keyed by `capability` (default = skill\n  name). Semantic duplicates under *different* names need an explicit shared\n  `capability:` in frontmatter (or content-similarity matching — out of scope,\n  design doc §Out of Scope). Hashes cover SKILL.md only (supporting files under\n  `references\u002F` and `scripts\u002F` are not hashed).\n- Snapshots accumulate under `memory\u002Fskill-graph-metrics\u002F`; they are the input\n  for future trend analysis (Phase 3, future work). Do not hand-edit them.\n","# Skill: skill-graph-analytics\n\nWeekly fleet analytics over the per-project skill-graph projections\n(`Projects\u002F\u003Cco-x>\u002Fdocs\u002Fskill-graph.json`, root `docs\u002Fskill-graph.json`).\nOne run answers: which skills does the fleet carry, which root skills never\nreached the projects, how far has each project drifted from the root\nprojection, and what is NEW or VANISHED fleet-wide since the last snapshot?\n\n**Triage only — never auto-modify.** This skill produces tickets and findings;\nit does not edit skills, promote content, or rewrite graphs.\n\n## When to Run\n\n- **Weekly cadence** — NO scheduled runner invokes this skill as of 2026-11-03\n  (automation roster: 01:30 fleet-review+resync, 04:00 governance-ticket —\n  neither carries the weekly consolidation; the old 02:30 reference here rotted,\n  T-20261101-004). **Revival home DECIDED 2026-11-03 (T-20261101-004)**: the\n  01:30 fleet runner's FRIDAY FULL OVERRIDE carries the weekly consolidation\n  (one prompt-line addition, no new schedule; runner prompt updated via\n  automation tooling). Until that override next fires, run it manually at the\n  start of a weekly maintenance session.\n- **On demand** — after a fleet-wide skill promotion\u002Fupgrade wave, before a\n  consolidation decision, or whenever skill-graph drift is suspected.\n\n## Step-by-Step Procedure\n\n### 1. Run the fleet report\n\n```bash\nbun scripts\u002Fskill-graph-fleet-report.ts\n```\n\nRead-only over all inputs. Writes one machine snapshot per local calendar date:\n`memory\u002Fskill-graph-metrics\u002Fsnapshot-\u003CYYYY-MM-DD>.json` (same-date reruns\noverwrite). First run establishes the baseline — no NEW\u002FVANISHED diff exists yet.\nProjects without a graph projection are skipped with a note; zero graphs found\nexits 1 (fix the generator, `bun scripts\u002Fgenerate-skill-graph.ts`, before triage).\n**Fleet membership SSOT** (DEC-20261107-01): the report's project set comes from\n`lib\u002Ffleet.ts` — `Projects\u002Fco-*` directories minus the exclusions recorded in\n`docs\u002Ffleet-membership.json` (verify with `bun scripts\u002Ffleet-list.ts --json`).\nNever derive the set with a raw `Projects\u002Fco-*` glob; convergence denominators\ncount members only.\n\n### 2. Read the snapshot and the report\n\n- **Per-project stats + Jaccard distance vs root** — bigger distance = narrower\n  or more divergent projection; small projects are expected to diverge, a large\n  project with rising distance is a delivery-gap signal.\n- **Presence matrix + top skills** — fleet presence counts per skill\n  **capability** (frontmatter `capability`, default = skill name; v1 project\n  graphs derive it from the bare node id, so v1 and v2 graphs join on the\n  same key).\n- **Skill convergence by capability** — capabilities carried by >= 3 project\n  graphs, split into `converged` (one content hash), `convergence-with-\n  divergence` (>= 2 hashes) and `hash-unknown` (a v1 project graph carries no\n  hashes yet); see §3 and §3.4.\n- **Root skills missing per project** — the delivery-drift surface.\n- **Root-graph orphan candidates** — graph-isolated root skills\u002Fagents with the\n  4-way cross-check axes (see §3.5); feeds the orphan triage rules.\n- **NEW\u002FVANISHED vs the previous snapshot** — fleet-wide skill evolution.\n\n### 3. Triage (judgment — not automatable)\n\n| Signal | Threshold | Action |\n|---|---|---|\n| **Capability** present in project graphs but **absent from the root graph**, `converged` class (single content hash) | >= 3 project graphs | Consolidation\u002Fpromotion candidate — file a ticket: `bun scripts\u002Fticket.ts create --manual \"skill-graph: \u003Ccapability> converged in N projects, absent from root — consolidation review\" --priority normal` (routes through §3.7.5 governance-backlog triage) |\n| Capability in the `convergence-with-divergence` class (>= 3 projects, >= 2 content hashes) | any | **Reconciliation candidate, not a promotion candidate** — see §3.4; never auto-promote |\n| Capability in the `hash-unknown` class | >= 3 project graphs | Treat as a promotion *candidate* by presence only; content equality is unverified until the carrying v1 projects are upgraded (`upgrade-project` delivers the v2 generator) — compare the SKILL.md files by hand before filing a promotion ticket |\n| **Root skill missing from many projects** | missing from >= 3 project graphs | Delivery-gap candidate — file a ticket naming the skill and the missing projects (likely template-delivery drift; check `templates\u002Fcommon\u002Fskills\u002F` and the variant contract) |\n| NEW\u002FVANISHED skills in the diff | any | Verify each is intentional (new skill landed \u002F skill deprecated) — unexplained entries are tickets |\n| **Root-graph orphan candidate** (report section \"Root-graph orphan candidates\") | any isolated skill\u002Fagent | Apply the 4-way orphan criteria below — never treat graph isolation alone as proof of orphanhood |\n| Everything else | — | No action; the snapshot is the record |\n\n### 3.4 Divergence triage (capability-based convergence)\n\nSince skill-graph v2 (ADR-0060 Amendment 11) a skill node's identity is\n`skill:\u003Cscope>\u002F\u003Cname>` with a `content_hash`, and fleet convergence groups by\n**capability** rather than by raw node id, so same-named skills whose content\ndiffers no longer look identical. For each `convergence-with-divergence`\ncapability in the report:\n\n1. Read the hash groups (`hash: projects`) and diff the SKILL.md copies of the\n   largest and smallest group.\n2. Classify: **fork** (intentional variant specialisation — set a distinct\n   `capability:` in the forked SKILL.md frontmatter, or keep and record why),\n   **drift** (accidental divergence — reconcile to one copy, then the group\n   becomes `converged` and a promotion candidate), or **rename** (same job,\n   different name — set the same `capability:` on both so they group).\n3. File one reconciliation ticket per capability (`bun scripts\u002Fticket.ts create\n   --manual \"skill-graph: \u003Ccapability> diverged across N projects — reconcile\"\n   --priority normal`); never edit the project copies from this skill.\n\nThe root `docs\u002Fskill-graph.md` section \"Same-Name Divergence (E2)\" and\n`bun scripts\u002Fverify-skill-graph.ts` carry the same signal inside the workspace\n(`version-drift` = same version, different content — bump or reconcile).\n`bun scripts\u002Fgenerate-skill-graph.ts --impact \u003Cskill>` lists the agents,\nprocedures, scopes, tests and contracts a change to a skill touches.\n\n### 3.5 Root-graph orphan cross-check (4-way criteria)\n\nA root skill or agent is an **orphan candidate** only when it is graph-isolated\n(zero edges — reported in the fleet report's orphan section since\nfleet-report v1.1.0). Isolation alone is NOT proof: classify each candidate by\ncrossing four axes, all reported in the same section:\n\n| Axis | Meaning | How the report shows it |\n|------|---------|------------------------|\n| **Definition** | The node exists in the graph (skills\u002F, agents\u002F) | the isolation list itself |\n| **Registry** | A SKILLS.md \u002F roster row exists | `registry ✓\u002F✗` |\n| **Mirror** | The 4 platform skill bases carry it | `mirrors N\u002F4` |\n| **Reference** | Workflow docs mention it (bounded corpus: platform\u002Fagent context docs + procedures\u002F + process\u002F) | `refs N` |\n\nTriage rules:\n\n- **All four axes present** (registry ✓, mirrors 4\u002F4, refs > 0) → *graph-edge\n  gap*, not an orphan. Fix the graph, not the skill: add a hand-reasoned\n  `docs\u002Fskill-graph.overrides.json` entry (`relates_to`, with `reason`\u002F`since`)\n  or a workflow-doc citation, then regenerate.\n- **Definition + Registry only** (no mirror, no refs) → *weak candidate* —\n  wire it into a workflow or schedule deprecation review.\n- **Missing registry row or mirrors** → *delivery drift* — the standard\n  registry\u002Fmirror fixes apply before any orphan verdict.\n- **Zero axes beyond definition** → *true orphan* — retire via the\n  skill\u002Fagent lifecycle manager.\n- Agent candidates: mirror axis does not apply (agents are not mirrored);\n  weigh roster presence (AGENTS.md §1\u002F§4.1) and doc references instead.\n\nHistorical baseline (2026-09-23 audit): 12 L0 + 48 variant skills were\ngraph-isolated; the workflow-doc citation scan (generator Source 4.8) plus four\noverrides entries brought L0 isolation to zero — variant-isolated skills remain\non this cadence's triage list.\n\n**Centrality-readiness note (Phase 2, future work)**: presence counts are a\nblunt proxy for importance. A later phase adds centrality ranking over the root\ngraph (degree\u002Fbetweenness over `step_uses_skill`\u002F`used_by` edges) per\ndocs\u002Fdesigns\u002F2026-09-22-skill-graph-fleet-analytics-design.md — do not\nover-prioritize a skill on presence alone today.\n\n### 4. Record findings\n\nAppend a `## Skill-Graph Fleet Analytics` section to the daily memory log\n(`memory\u002FYYYY-MM-DD.md`): snapshot path, headline stats (projects covered,\nfleet skill count, top-3 by presence), every ticket filed, and any\ndelivery-gap or convergence observation that did not merit a ticket.\n\n## Notes\n\n- Capability-keyed presence: presence is keyed by `capability` (default = skill\n  name). Semantic duplicates under *different* names need an explicit shared\n  `capability:` in frontmatter (or content-similarity matching — out of scope,\n  design doc §Out of Scope). Hashes cover SKILL.md only (supporting files under\n  `references\u002F` and `scripts\u002F` are not hashed).\n- Snapshots accumulate under `memory\u002Fskill-graph-metrics\u002F`; they are the input\n  for future trend analysis (Phase 3, future work). Do not hand-edit them.\n",{"name":5,"owner":23,"scope":24,"status":25,"version":26,"metadata":27,"relates_to":37,"description":44,"l2_propagate":19,"last_reviewed":45,"prerequisites":46},"pm","workspace","active","1.4.0",{"type":28,"triggers":29},"process",[30,31,32,33,34,35,36],"skill-graph analytics","fleet skill graph","skill graph report","skill graph fleet report","skill convergence triage","orphan agent check","orphan skill check",[38,41],{"type":39,"skill":40},"composes_with","project-resync",{"type":42,"skill":43},"relates_to","context-commonization-review","Weekly fleet analytics over the per-project skill-graph projections: consolidate root + Projects\u002F\u003Cco-x> graphs into a dated snapshot, triage skill convergence (consolidation\u002Fpromotion candidates), delivery drift (root skills missing from projects), and root-graph orphans (4-way cross-check), and record findings. Use when: the weekly analytics cadence fires (manual until the Friday full-override revival — see When to Run), the user says \"skill-graph analytics\", \"fleet skill graph\", \"skill graph report\", or asks which skills the project fleet carries, which root skills never reached projects, whether any agent or skill is orphaned, or what changed fleet-wide since last week.\n",1794096000000,"workspace-root CWD; docs\u002Fskill-graph.json + Projects\u002F\u003Cco-x>\u002Fdocs\u002Fskill-graph.json projections present","main","1162ccc568cf90c01b3886d6143cbfd851a37883","templates\u002Fcommon\u002F.claude\u002Fskills\u002Fskill-graph-analytics",9,"https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards","https:\u002F\u002Fgithub.com\u002F5throck","Experienced in digital transformation, enterprise IT, startup acceleration, and business strategy. Focused on building practical systems that connect technolo","2026-11-14 21:22:34",{"source":8,"skillName":5,"commands":56,"prompt":66,"zipAvailable":67},[57,60,63],{"tool":58,"command":59},"npx","npx skills add 5throck\u002Fai-workspace-standards --skill skill-graph-analytics",{"tool":61,"command":62},"bunx","bunx skills add 5throck\u002Fai-workspace-standards --skill skill-graph-analytics",{"tool":64,"command":65},"pnpm","pnpm dlx skills add 5throck\u002Fai-workspace-standards --skill skill-graph-analytics","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fcommon\u002F.claude\u002Fskills\u002Fskill-graph-analytics\u002FSKILL.md",true,[],[70,78,83,88,93,98],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":77},"01M7BBJDYHJMGTJYSC37JVWY8S","i18n-layout","Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL\u002Fbidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF\u002FBOM issues surface.","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":79,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":77},"01M7BBJ9XHAAS1AVGKZZ841VBZ","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.hermes\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":84,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":77},"01M7BBJ5C3X8TKFKG5R9P9P49W","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.gemini\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":89,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":77},"01M7BBJ1B52X0QMJ2QVBMPJB37","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.codex\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":94,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":77},"01M7BBHX91BW2Q2R276MRX5AW3","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.claude\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":99,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":77},"01M7BBHQWGXPSFRWN1PRC2NMMV","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.agents\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[]]