[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M758ZB3XF7THZPQYR4TBD1W8":3,"skill-same:01M758ZB3XF7THZPQYR4TBD1W8":70},{"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":26,"commitSha":27,"skillDirectory":28,"forksCount":14,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M758ZB3XF7THZPQYR4TBD1W8","ag-context-discovery",null,"Discover and load all relevant context for the current task. Lists feature group nested files with full paths, loads process\u002Fcontext\u002F files by domain routing. Called at the start of every agent session.","ngxccc\u002Fagent-skills-kit","ngxccc","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F163379246?v=4","01M758XVG64QK26AGK0WKB26TB","Ngoc Tran","01M758XVHB8J6MHV0YTBVDBXCA",0,"MIT","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fag-context-discovery\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: ag-context-discovery\ndescription: \"Discover and load all relevant context for the current task. Lists feature group nested files with full paths, loads process\u002Fcontext\u002F files by domain routing. Called at the start of every agent session.\"\ntrigger_keywords: context discovery, load context, feature folder files, context routing discovery\nlayer: contract\n---\n\n# ag-context-discovery\n\n## When to Apply\n\nUse this skill when working with ag-context-discovery workflows, tasks, or system specifications.\n\n## How to Use\n\nRefer to the workflow instructions and command references detailed below.\n\n> **Output style:** Follow `process\u002Fdevelopment-protocols\u002Fcommunication-standards.md` — answer-first, plain language, no unexplained jargon, TL;DR on long responses.\n\nDiscover and load all relevant context for the current task. This skill lists feature group nested files with full paths and loads `process\u002Fcontext\u002F` files by domain routing table. It is the canonical context-loading entrypoint for every agent session.\n\n## When To Invoke\n\nAt the start of every agent session (research, innovate, plan, validate, execute, update-process, fast-mode). Also at the start of any skill that needs repo context before operating.\n\n## Frontmatter Schemas\n\nDocument the canonical schemas for three file types used across the repo. These schemas enable frontmatter-aware routing and filtering.\n\n**Context file frontmatter schema:**\n\n```yaml\nname: context:{slug}\ndescription: \"one-line scope — used by ag-context-discovery for routing\"\nkeywords: comma, separated, task, vocabulary, terms # drives grep-first keyword routing\nrelated: [context: { other-slug }, context: { another-slug }] # sibling\u002Fcross-group links (optional)\ndate: dd-mm-yy\n```\n\n- `keywords` — strongly recommended, non-empty. Comma-separated task-vocabulary terms an\n  agent would use to describe a task this doc serves (e.g. `session, token, refresh, jwt,\nlogin`). This is the match surface for `discover-context.mjs --match`; weak\u002Fabsent\n  keywords are why a relevant doc never gets routed to. Lint WARNS when empty (so existing\n  projects don't break on sync) — backfill at UPDATE-PROCESS.\n- `related` — OPTIONAL list of `context:{slug}` values for sibling docs that are usually\n  needed together (the markdown-native equivalent of cross-links). Every slug listed MUST\n  resolve to a real `context:` doc; dangling links fail lint. Discovery follows these after\n  the primary match so a task touching two domains loads both.\n\n**Plan file frontmatter schema:**\n\n```yaml\nname: plan:{slug}\ndescription: \"one-line scope and feature\"\ndate: dd-mm-yy\nfeature: { feature-folder-name }\nphase: \"{phase-id}\" # optional, for phase program plans only\n```\n\n**Report file frontmatter schema:**\n\n```yaml\nname: report:{slug}\ndescription: \"one-line scope\"\ndate: dd-mm-yy\nmetadata:\n  node_type: memory\n  type: report\n  feature: { feature-folder-name }\n  phase: { phase-id }\n```\n\n## Invocation\n\n**Primary method** — run the auto-discovery script. It lists all nested files under\n`process\u002Fcontext\u002F`, `process\u002Fdevelopment-protocols\u002F`, `process\u002Fgeneral-plans\u002Factive\u002F`,\nand (with `--feature`) the feature folder, extracting ONLY the leading YAML frontmatter\nblock of each `.md` file (no whole-file reads):\n\n```bash\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs [--feature \u003Cname>] [--json]\n```\n\nThe script groups output into: context files with frontmatter, protocol files, feature\nfiles by subfolder, active general plans, and files-without-frontmatter (path only). It\nnever throws on a missing root and exits 0 unless given a bad flag. Use `--json` for a\nmachine-readable object. Prefer this over manually reading each file — it is deterministic\nand avoids loading huge files into context.\n\n**Keyword-first routing (deterministic, not judgment).** When the task vocabulary does not\nobviously map to a routing-table row, do not guess — let the index do the matching:\n\n```bash\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs --match \"update the user ORM model\"\n```\n\nThis tokenizes the task and ranks context docs by overlap with their frontmatter `keywords`,\nthen appends any `related:` siblings of the top hits. Read the ranked docs in order. This is\nthe fallback that fixes \"the right doc existed but the agent walked past it.\"\n\n**Routing-table generation (drift-proof index).** The \"Current Root Entry Points\" and\n\"Current Context Groups\" tables in `all-context.md` are GENERATED from frontmatter, not\nhand-authored — they live between `\u003C!-- GENERATED:routing -->` \u002F `\u003C!-- \u002FGENERATED:routing -->`\nmarkers. Rebuild them after any context-org change:\n\n```bash\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs --emit-routing   # rewrites the block\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs --check-routing  # lint: block in sync?\n```\n\nThe hand-authored Task Routing Table (task-type → file) stays editorial and is NOT generated.\n\nPer **task-folder artefact colocation**, the script surfaces each task's plan, spec,\nreports, and references INSIDE its own `{slug}_{date}\u002F` task folder; the sibling\n`reports\u002F`\u002F`references\u002F` dirs are deprecated and only hold legacy artefacts.\n\n## Workflow (manual FALLBACK)\n\nUse these steps only if the script above fails or is unavailable.\n\n**Step 1.** Run `find process\u002Fcontext\u002F -type f | sort` and record all available context files.\n\n**Step 2.** Run `find process\u002Fdevelopment-protocols\u002F -type f | sort` and record all protocol files.\n\n**Step 3.** Read `process\u002Fcontext\u002Fall-context.md` to get the routing table and current feature list.\n\n**Step 4.** If a feature name was provided as the argument, run `find process\u002Ffeatures\u002F{feature}\u002F -type f | sort` to list ALL artifacts across all subfolders (`active\u002F`, `completed\u002F`, `backlog\u002F`, plus any legacy `reports\u002F`, `references\u002F`). Surface full file paths — not just folder names. Per **task-folder artefact colocation**, expect each task's plan, spec, reports, and references INSIDE its own `{slug}_{date}\u002F` task folder; the sibling `reports\u002F`\u002F`references\u002F` dirs are deprecated and only hold legacy artefacts.\n\n**Step 5.** If no feature name was provided, run `find process\u002Fgeneral-plans\u002Factive\u002F -type f | sort` to surface any active plans relevant to the current task. Note: plan files are inside `{slug}_{date}\u002F` task subfolders — look one level deep for `*_PLAN_*.md` files.\n\n**Step 6.** From the routing table in `all-context.md`, identify the context group files relevant to the current task domain (e.g. `tests`, `container`, `infra`, `skills`, `uxui`, `workflows`). Do NOT read every context file — only the ones the routing table says apply to this domain. If no row obviously matches, run `discover-context.mjs --match \"\u003Ctask>\"` and use its ranked keyword hits instead of guessing.\n\n**Step 7.** Read the domain context file(s) identified in step 6. Each `all-{group}.md` is a router — after reading it, follow its routing table to load the deeper domain file(s) for the task. Then load any `related:` siblings of the docs you read — a task spanning two domains (e.g. auth + tests) needs both, and `related` is how the cross-domain link is declared.\n\n**Step 8.** Frontmatter extraction: For each file in the discovered set, if the file has YAML frontmatter (a `---` block at top), extract the `name`, `description`, and `date` fields. Surface in output as `\"[path] (name: X — description: Y)\"`. Files without frontmatter: surface path only (no error, do not skip them).\n\n**Step 9.** Report: full file listing for the feature folder (if applicable), active context files loaded, and any open gaps in context that would block the task.\n\n## Output Format\n\n```\nContext file paths with frontmatter:\n  process\u002Fcontext\u002Fskills\u002Fskill-apps.md (context:skill-apps — Skill app runtime, vite architecture, ctx-gateway, deployment)\n  process\u002Fcontext\u002Ftests\u002Fall-tests.md (context:all-tests — Test routing, runner split, debugging procedures)\n\nPlan files with frontmatter:\n  process\u002Ffeatures\u002F{feature}\u002Fcompleted\u002F{slug}_{date}\u002F{slug}_PLAN_{date}.md (plan:{plan-name} — short description...)\n\nFiles without frontmatter (path only):\n  process\u002Fcontext\u002Fui\u002Fdesign.md\n```\n\n## Context Envelope\n\nAt session start, every inner-loop agent (research \u002F plan \u002F execute \u002F update-process) emits a\n**Context Envelope** — a 10-field table capturing the orientation an agent needs to act. All 10\nfields are required (use best-effort values; `TBD — [reason]` when not yet determinable). The fields\nMUST appear in the EXACT canonical C-2 order below — identical order in this SKILL and in all four\ninner-loop agents:\n\n| #   | Field                   | Value                                                                                                 |\n| --- | ----------------------- | ----------------------------------------------------------------------------------------------------- |\n| 1   | `feature`               | feature folder name (or `TBD`)                                                                        |\n| 2   | `phase`                 | current RIPER phase (`RESEARCH` \u002F `INNOVATE` \u002F `PLAN` \u002F `PVL` \u002F `EXECUTE` \u002F `EVL` \u002F `UPDATE-PROCESS`) |\n| 3   | `session-goal`          | one-line goal from the `\u002Fgoal` block                                                                  |\n| 4   | `branch`                | current git branch                                                                                    |\n| 5   | `worktree`              | worktree path (or `main`)                                                                             |\n| 6   | `context-group`         | relevant `process\u002Fcontext\u002F` group (or `none`)                                                         |\n| 7   | `blast-radius-packages` | packages\u002Fpaths in scope (comma-separated or `TBD`)                                                    |\n| 8   | `active-plan`           | selected plan file path (or `none`)                                                                   |\n| 9   | `test-runner`           | test runner(s); multi-runner uses pipe-delimited DISPLAY format `bun test \\| vitest`                  |\n| 10  | `validate-contract`     | validate-contract path (or `none`)                                                                    |\n\n**Canonical order (memorize):** `feature → phase → session-goal → branch → worktree → context-group\n→ blast-radius-packages → active-plan → test-runner → validate-contract`.\n\n**`test-runner` multi-runner rule:** the pipe-delimited `bun test | vitest` value is a DISPLAY\nconvention only. The phase-loop workflow template\n(`.claude\u002Fskills\u002Fag-generate-phase-program\u002Ftemplates\u002Fphase-loop-workflow-template.js`) expands it into\nSEQUENTIAL test steps (`bun test` THEN `vitest`) — a literal `bun test | vitest` shell pipe is NEVER\nemitted or run. See `03-session-start.md` for the matching field table.\n\n## Frontmatter-Aware Routing\n\nAfter collecting file paths, read YAML frontmatter from each file where present.\n\nSurface the following fields alongside path: `name`, `description`, `keywords`, `related`, `date`, `type`, `feature`, `phase`.\n\nUse the `description` and `keywords` fields for routing decisions instead of filename inference.\n\nGroup plan files by their `feature` field value.\n\nFilter and sort by `type` field (`context` \u002F `plan` \u002F `report` \u002F `references`).\n\nDiscover all nested files under feature group subdirs: `active\u002F`, `completed\u002F`, `backlog\u002F`, `reports\u002F`, `references\u002F`.\n\nFiles without frontmatter: surface path only (no error), do not skip them.\n\n## Important Rules\n\n- This skill is **INDEPENDENT** — it does not invoke other skills.\n- Read `all-context.md` as a router, then load the deeper file(s). Do not treat `all-context.md` as sufficient on its own.\n- Always produce the full `find` output, not a summary. The exact file paths are the output.\n- Never hardcode context file paths — always discover via the `find` command and the routing table in `all-context.md`.\n\n## References\n\n- [process\u002Fcontext\u002Fall-context.md](process\u002Fcontext\u002Fall-context.md)\n","# ag-context-discovery\n\n## When to Apply\n\nUse this skill when working with ag-context-discovery workflows, tasks, or system specifications.\n\n## How to Use\n\nRefer to the workflow instructions and command references detailed below.\n\n> **Output style:** Follow `process\u002Fdevelopment-protocols\u002Fcommunication-standards.md` — answer-first, plain language, no unexplained jargon, TL;DR on long responses.\n\nDiscover and load all relevant context for the current task. This skill lists feature group nested files with full paths and loads `process\u002Fcontext\u002F` files by domain routing table. It is the canonical context-loading entrypoint for every agent session.\n\n## When To Invoke\n\nAt the start of every agent session (research, innovate, plan, validate, execute, update-process, fast-mode). Also at the start of any skill that needs repo context before operating.\n\n## Frontmatter Schemas\n\nDocument the canonical schemas for three file types used across the repo. These schemas enable frontmatter-aware routing and filtering.\n\n**Context file frontmatter schema:**\n\n```yaml\nname: context:{slug}\ndescription: \"one-line scope — used by ag-context-discovery for routing\"\nkeywords: comma, separated, task, vocabulary, terms # drives grep-first keyword routing\nrelated: [context: { other-slug }, context: { another-slug }] # sibling\u002Fcross-group links (optional)\ndate: dd-mm-yy\n```\n\n- `keywords` — strongly recommended, non-empty. Comma-separated task-vocabulary terms an\n  agent would use to describe a task this doc serves (e.g. `session, token, refresh, jwt,\nlogin`). This is the match surface for `discover-context.mjs --match`; weak\u002Fabsent\n  keywords are why a relevant doc never gets routed to. Lint WARNS when empty (so existing\n  projects don't break on sync) — backfill at UPDATE-PROCESS.\n- `related` — OPTIONAL list of `context:{slug}` values for sibling docs that are usually\n  needed together (the markdown-native equivalent of cross-links). Every slug listed MUST\n  resolve to a real `context:` doc; dangling links fail lint. Discovery follows these after\n  the primary match so a task touching two domains loads both.\n\n**Plan file frontmatter schema:**\n\n```yaml\nname: plan:{slug}\ndescription: \"one-line scope and feature\"\ndate: dd-mm-yy\nfeature: { feature-folder-name }\nphase: \"{phase-id}\" # optional, for phase program plans only\n```\n\n**Report file frontmatter schema:**\n\n```yaml\nname: report:{slug}\ndescription: \"one-line scope\"\ndate: dd-mm-yy\nmetadata:\n  node_type: memory\n  type: report\n  feature: { feature-folder-name }\n  phase: { phase-id }\n```\n\n## Invocation\n\n**Primary method** — run the auto-discovery script. It lists all nested files under\n`process\u002Fcontext\u002F`, `process\u002Fdevelopment-protocols\u002F`, `process\u002Fgeneral-plans\u002Factive\u002F`,\nand (with `--feature`) the feature folder, extracting ONLY the leading YAML frontmatter\nblock of each `.md` file (no whole-file reads):\n\n```bash\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs [--feature \u003Cname>] [--json]\n```\n\nThe script groups output into: context files with frontmatter, protocol files, feature\nfiles by subfolder, active general plans, and files-without-frontmatter (path only). It\nnever throws on a missing root and exits 0 unless given a bad flag. Use `--json` for a\nmachine-readable object. Prefer this over manually reading each file — it is deterministic\nand avoids loading huge files into context.\n\n**Keyword-first routing (deterministic, not judgment).** When the task vocabulary does not\nobviously map to a routing-table row, do not guess — let the index do the matching:\n\n```bash\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs --match \"update the user ORM model\"\n```\n\nThis tokenizes the task and ranks context docs by overlap with their frontmatter `keywords`,\nthen appends any `related:` siblings of the top hits. Read the ranked docs in order. This is\nthe fallback that fixes \"the right doc existed but the agent walked past it.\"\n\n**Routing-table generation (drift-proof index).** The \"Current Root Entry Points\" and\n\"Current Context Groups\" tables in `all-context.md` are GENERATED from frontmatter, not\nhand-authored — they live between `\u003C!-- GENERATED:routing -->` \u002F `\u003C!-- \u002FGENERATED:routing -->`\nmarkers. Rebuild them after any context-org change:\n\n```bash\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs --emit-routing   # rewrites the block\nnode .claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs --check-routing  # lint: block in sync?\n```\n\nThe hand-authored Task Routing Table (task-type → file) stays editorial and is NOT generated.\n\nPer **task-folder artefact colocation**, the script surfaces each task's plan, spec,\nreports, and references INSIDE its own `{slug}_{date}\u002F` task folder; the sibling\n`reports\u002F`\u002F`references\u002F` dirs are deprecated and only hold legacy artefacts.\n\n## Workflow (manual FALLBACK)\n\nUse these steps only if the script above fails or is unavailable.\n\n**Step 1.** Run `find process\u002Fcontext\u002F -type f | sort` and record all available context files.\n\n**Step 2.** Run `find process\u002Fdevelopment-protocols\u002F -type f | sort` and record all protocol files.\n\n**Step 3.** Read `process\u002Fcontext\u002Fall-context.md` to get the routing table and current feature list.\n\n**Step 4.** If a feature name was provided as the argument, run `find process\u002Ffeatures\u002F{feature}\u002F -type f | sort` to list ALL artifacts across all subfolders (`active\u002F`, `completed\u002F`, `backlog\u002F`, plus any legacy `reports\u002F`, `references\u002F`). Surface full file paths — not just folder names. Per **task-folder artefact colocation**, expect each task's plan, spec, reports, and references INSIDE its own `{slug}_{date}\u002F` task folder; the sibling `reports\u002F`\u002F`references\u002F` dirs are deprecated and only hold legacy artefacts.\n\n**Step 5.** If no feature name was provided, run `find process\u002Fgeneral-plans\u002Factive\u002F -type f | sort` to surface any active plans relevant to the current task. Note: plan files are inside `{slug}_{date}\u002F` task subfolders — look one level deep for `*_PLAN_*.md` files.\n\n**Step 6.** From the routing table in `all-context.md`, identify the context group files relevant to the current task domain (e.g. `tests`, `container`, `infra`, `skills`, `uxui`, `workflows`). Do NOT read every context file — only the ones the routing table says apply to this domain. If no row obviously matches, run `discover-context.mjs --match \"\u003Ctask>\"` and use its ranked keyword hits instead of guessing.\n\n**Step 7.** Read the domain context file(s) identified in step 6. Each `all-{group}.md` is a router — after reading it, follow its routing table to load the deeper domain file(s) for the task. Then load any `related:` siblings of the docs you read — a task spanning two domains (e.g. auth + tests) needs both, and `related` is how the cross-domain link is declared.\n\n**Step 8.** Frontmatter extraction: For each file in the discovered set, if the file has YAML frontmatter (a `---` block at top), extract the `name`, `description`, and `date` fields. Surface in output as `\"[path] (name: X — description: Y)\"`. Files without frontmatter: surface path only (no error, do not skip them).\n\n**Step 9.** Report: full file listing for the feature folder (if applicable), active context files loaded, and any open gaps in context that would block the task.\n\n## Output Format\n\n```\nContext file paths with frontmatter:\n  process\u002Fcontext\u002Fskills\u002Fskill-apps.md (context:skill-apps — Skill app runtime, vite architecture, ctx-gateway, deployment)\n  process\u002Fcontext\u002Ftests\u002Fall-tests.md (context:all-tests — Test routing, runner split, debugging procedures)\n\nPlan files with frontmatter:\n  process\u002Ffeatures\u002F{feature}\u002Fcompleted\u002F{slug}_{date}\u002F{slug}_PLAN_{date}.md (plan:{plan-name} — short description...)\n\nFiles without frontmatter (path only):\n  process\u002Fcontext\u002Fui\u002Fdesign.md\n```\n\n## Context Envelope\n\nAt session start, every inner-loop agent (research \u002F plan \u002F execute \u002F update-process) emits a\n**Context Envelope** — a 10-field table capturing the orientation an agent needs to act. All 10\nfields are required (use best-effort values; `TBD — [reason]` when not yet determinable). The fields\nMUST appear in the EXACT canonical C-2 order below — identical order in this SKILL and in all four\ninner-loop agents:\n\n| #   | Field                   | Value                                                                                                 |\n| --- | ----------------------- | ----------------------------------------------------------------------------------------------------- |\n| 1   | `feature`               | feature folder name (or `TBD`)                                                                        |\n| 2   | `phase`                 | current RIPER phase (`RESEARCH` \u002F `INNOVATE` \u002F `PLAN` \u002F `PVL` \u002F `EXECUTE` \u002F `EVL` \u002F `UPDATE-PROCESS`) |\n| 3   | `session-goal`          | one-line goal from the `\u002Fgoal` block                                                                  |\n| 4   | `branch`                | current git branch                                                                                    |\n| 5   | `worktree`              | worktree path (or `main`)                                                                             |\n| 6   | `context-group`         | relevant `process\u002Fcontext\u002F` group (or `none`)                                                         |\n| 7   | `blast-radius-packages` | packages\u002Fpaths in scope (comma-separated or `TBD`)                                                    |\n| 8   | `active-plan`           | selected plan file path (or `none`)                                                                   |\n| 9   | `test-runner`           | test runner(s); multi-runner uses pipe-delimited DISPLAY format `bun test \\| vitest`                  |\n| 10  | `validate-contract`     | validate-contract path (or `none`)                                                                    |\n\n**Canonical order (memorize):** `feature → phase → session-goal → branch → worktree → context-group\n→ blast-radius-packages → active-plan → test-runner → validate-contract`.\n\n**`test-runner` multi-runner rule:** the pipe-delimited `bun test | vitest` value is a DISPLAY\nconvention only. The phase-loop workflow template\n(`.claude\u002Fskills\u002Fag-generate-phase-program\u002Ftemplates\u002Fphase-loop-workflow-template.js`) expands it into\nSEQUENTIAL test steps (`bun test` THEN `vitest`) — a literal `bun test | vitest` shell pipe is NEVER\nemitted or run. See `03-session-start.md` for the matching field table.\n\n## Frontmatter-Aware Routing\n\nAfter collecting file paths, read YAML frontmatter from each file where present.\n\nSurface the following fields alongside path: `name`, `description`, `keywords`, `related`, `date`, `type`, `feature`, `phase`.\n\nUse the `description` and `keywords` fields for routing decisions instead of filename inference.\n\nGroup plan files by their `feature` field value.\n\nFilter and sort by `type` field (`context` \u002F `plan` \u002F `report` \u002F `references`).\n\nDiscover all nested files under feature group subdirs: `active\u002F`, `completed\u002F`, `backlog\u002F`, `reports\u002F`, `references\u002F`.\n\nFiles without frontmatter: surface path only (no error), do not skip them.\n\n## Important Rules\n\n- This skill is **INDEPENDENT** — it does not invoke other skills.\n- Read `all-context.md` as a router, then load the deeper file(s). Do not treat `all-context.md` as sufficient on its own.\n- Always produce the full `find` output, not a summary. The exact file paths are the output.\n- Never hardcode context file paths — always discover via the `find` command and the routing table in `all-context.md`.\n\n## References\n\n- [process\u002Fcontext\u002Fall-context.md](process\u002Fcontext\u002Fall-context.md)\n",{"name":5,"layer":24,"description":7,"trigger_keywords":25},"contract","context discovery, load context, feature folder files, context routing discovery","main","1ef0dcd00fe806a8dc4e0b7173b9db967c3ab274",".claude\u002Fskills\u002Fag-context-discovery","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit","https:\u002F\u002Fgithub.com\u002Fngxccc","https:\u002F\u002Fngxccc.github.io","Make it work, make it right, make it fast.","2026-08-02 20:59:07",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add ngxccc\u002Fagent-skills-kit --skill ag-context-discovery",{"tool":40,"command":41},"bunx","bunx skills add ngxccc\u002Fagent-skills-kit --skill ag-context-discovery",{"tool":43,"command":44},"pnpm","pnpm dlx skills add ngxccc\u002Fagent-skills-kit --skill ag-context-discovery","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fag-context-discovery\u002FSKILL.md",true,[48,54,58,62,66],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"scripts\u002Fdiscover-context.mjs","SCRIPT","脚本",13232,"https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-context.mjs",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"scripts\u002Fdiscover-skills.mjs",3233,"https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fdiscover-skills.mjs",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Ffixtures\u002Fvalidate-context-discovery-output\u002Ffail.md",791,"https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Ffixtures\u002Fvalidate-context-discovery-output\u002Ffail.md",{"path":63,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":64,"sourceHtmlUrl":65},"scripts\u002Ffixtures\u002Fvalidate-context-discovery-output\u002Fpass.md",827,"https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Ffixtures\u002Fvalidate-context-discovery-output\u002Fpass.md",{"path":67,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":68,"sourceHtmlUrl":69},"scripts\u002Fvalidate-context-discovery-output.mjs",2909,"https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fag-context-discovery\u002Fscripts\u002Fvalidate-context-discovery-output.mjs",[71,78,85,92,99,106],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M76W3C8X6Q5DS0364KQT7W7A","zustand","Rules and standards for state management using Zustand in React and Next.js. Use when implementing client-side stores, persistence with localStorage, or preventing SSR hydration mismatch.","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002Fskills\u002F.curated\u002Fzustand\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":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M76W39FJGES309VAH7SFWA9Y","zod","Zod v4 schema validation best practices. Use when defining schemas, utilizing safeParse, or customizing error messages.","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002Fskills\u002F.curated\u002Fzod\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M76W36HTYA4BR5FKCWMX1RFR","upstash-ratelimit","Rules, standards, and references for rate limiting in serverless environments using Upstash Ratelimit. Use when implementing rate limiting, configuring Upstash Redis, or handling request blocking.","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002Fskills\u002F.curated\u002Fupstash-ratelimit\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M76W32H4PS01WFRS7FNPK2G1","ui-design","UI\u002FUX and frontend design best practices guidelines (formerly frontend-design). This skill should be used when writing, reviewing, or designing frontend code to ensure accessibility, performance, and usability. Triggers on tasks involving HTML structure, CSS styling, responsive layouts, form design, animations, or accessibility improvements.","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002Fskills\u002F.curated\u002Fui-design\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":19},"01M76W2YAP6RK8N0EVWFG54VD7","typescript","TypeScript performance, tsconfig, type errors, async patterns — triggered when the user asks to \"optimize TypeScript performance\", \"speed up tsc compilation\", \"configure tsconfig.json\", \"fix type errors\", \"improve async patterns\", or encounters TS errors (TS2322, TS2339, \"is not assignable to\"). Also triggers on .ts, .tsx, .d.ts file work involving type definitions, module organization, or memory management. Does NOT cover TypeScript basics, framework-specific patterns, or testing.","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002Fskills\u002F.curated\u002Ftypescript\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":19},"01M76W2RMAZNNJQ2HECH3X6ZSH","typescript-refactor","TypeScript and TSX refactoring and modernization guidelines from a principal specialist perspective, current to TypeScript 6.0 and React 19. This skill should be used when refactoring, reviewing, or modernizing TypeScript or React\u002FTSX code for type safety, compiler performance, and idiomatic patterns. Triggers on tasks involving type architecture, narrowing, generics, discriminated unions, error handling, React component and hook typing, or migration to modern TypeScript features (satisfies, using, const type parameters, inferred type predicates, isolatedDeclarations, erasable syntax, import attributes).","https:\u002F\u002Fgithub.com\u002Fngxccc\u002Fagent-skills-kit\u002Fblob\u002Fmain\u002Fskills\u002F.curated\u002Ftypescript-refactor\u002FSKILL.md",[],[]]