[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71RE3RGGT2QEKZFY7NMXT78":3,"skill-same:01M71RE3RGGT2QEKZFY7NMXT78":116},{"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":28,"publishedAt":49,"linkOnly":50,"markdownContent":51,"markdownBody":52,"frontMatter":53,"refName":54,"commitSha":55,"skillDirectory":56,"forksCount":57,"repositoryHtmlUrl":58,"repositoryHomepageUrl":6,"ownerHtmlUrl":59,"ownerBlogUrl":6,"ownerDescription":60,"updatedAt":61,"install":62,"resources":75},"01M71RE3RGGT2QEKZFY7NMXT78","autospec-design",null,"Use when the user wants to adopt a vendor design language for the current repo — fetches the per-vendor DESIGN.md from the berlinguyinca\u002Fawesome-design-md catalog, scores the repo against candidate vendors, writes DESIGN.md at the project root, and optionally hands off a migration spec to \u002Fautospec-define. Subcommands suggest, apply, and migrate.","berlinguyinca\u002Fautospec","berlinguyinca","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F16364?v=4","01M71R7C0SWBG05FNHQE4XQ7Y3","Gert Wohlgemuth","01M71R7C1VY5K3Q715RNDDBVGQ",3,"MIT","https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48],"autospec","autospec design","design","migration","spec","at","vendor","design-md","adopt","design language","candidate","migrate","apply","when","use","define","catalog","repo","project","for","2026-11-10 13:02:02",false,"---\nname: autospec-design\ndescription: Use when the user wants to adopt a vendor design language for the current repo — fetches the per-vendor DESIGN.md from the berlinguyinca\u002Fawesome-design-md catalog, scores the repo against candidate vendors, writes DESIGN.md at the project root, and optionally hands off a migration spec to \u002Fautospec-define. Subcommands suggest, apply, and migrate.\n---\n\n# autospec-design (harness-neutral)\n\nSkill scaffold for the **autospec-design** workflow. The full workflow lets the\nuser pick a vendor design language (Linear, Stripe, Vercel, ...) from a curated\ncatalog, write `DESIGN.md` at the project root, and optionally hand off a\nmigration spec into `\u002Fautospec-define`. The catalog lives at\n`berlinguyinca\u002Fawesome-design-md` and is fetched at runtime — never vendored.\n\nThis file is the **autospec-design trio body**. The `suggest` and `apply`\nsubcommands are implemented; `migrate` still holds a placeholder until issue\n#580 lands.\n\nManage your own context — never exceed 60%. Delegate to subagents whenever your harness supports it.\n\n\u003C!-- autospec-block:startup-self-update SKILL_NAME=autospec-design -->\n\n## Self-update mode\n\nIf the feature-request argument matches the regex `^\\s*update\\s*$` (case-insensitive, whitespace-padded), this skill enters self-update mode and does not run the normal pipeline:\n\n1. **Detect harness** by checking which install path exists for this skill:\n   - Claude Code: `~\u002F.claude\u002Fskills\u002Fautospec-design\u002FSKILL.md`\n   - OpenCode:    `~\u002F.config\u002Fopencode\u002Fagent\u002Fautospec-design.md`\n   - Codex CLI:   `~\u002F.codex\u002Fprompts\u002Fautospec-design.md`\n2. **Re-install the full autospec suite from `main`** by piping the canonical installer:\n   ```bash\n   curl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002Fberlinguyinca\u002Fautospec\u002Fmain\u002Fbootstrap.sh | bash -s -- --skill all --harness all --update\n   ```\n   Run this one-liner once; it refreshes all autospec skills across all harnesses.\n3. **Show the diff** between the prior installed file(s) and the freshly fetched copy (e.g. `diff \u003C(cat \u003Cprior>) \u003C(curl -fsSL ...SKILL.md)` or the equivalent recorded by the installer).\n4. **Stop.** Do not enter any subcommand. Print the upgrade summary and return to the user.\n\nIf no install path is detected, print `Self-update: no installed copy of autospec-design found; run install.sh first.` and exit.\n\n## Invocation\n\n```\n\u002Fautospec-design suggest\n\u002Fautospec-design apply \u003Cvendor> [--force] [--branch \u003Cname>]\n\u002Fautospec-design migrate \u003Cvendor> [--branch \u003Cname>]\n```\n\n- `suggest` — scan the current repo, score every catalog vendor by the\n  framework \u002F domain \u002F brand-language rubric, present the top 3 with a\n  one-line rationale each. Prints only — never modifies any files.\n- `apply \u003Cvendor>` — fetch `\u003Cvendor>\u002FDESIGN.md` from the catalog, write it\n  to `\u003Crepo-root>\u002FDESIGN.md` on a fresh feature branch (default:\n  `feat\u002Fdesign-\u003Cvendor>`), commit, and push. Refuses to overwrite an\n  existing `DESIGN.md` without `--force`. Does **not** open a PR.\n- `migrate \u003Cvendor>` — generate a migration spec at\n  `docs\u002Fspecs\u002F\u003CYYYY-MM-DD>-design-migration-\u003Cvendor>.md` from the scanned\n  UI inventory and hand off to `\u002Fautospec-define \u003Cspec-path>`.\n\nThe skill exits non-zero if `gh auth status` fails when issuing a network\noperation that needs authentication. See spec\n`docs\u002Fspecs\u002F2026-05-26-autospec-design-skill.md` for the full contract.\n\n## Required capabilities & harness adapter\n\nThis workflow assumes a small set of capabilities. Map each one to your harness's actual tool. If a capability is missing, use the listed fallback.\n\n| Capability                  | Claude Code                          | OpenCode                                 | Codex CLI                                | Fallback if missing                                |\n|-----------------------------|--------------------------------------|------------------------------------------|------------------------------------------|----------------------------------------------------|\n| Read-only codebase research | `Agent` (subagent_type=Explore)      | `task` agent in read-only mode           | `apply_patch` read-only \u002F shell `grep`   | Do the search in-thread with `rg`\u002F`grep`           |\n| Foreground delegation       | `Agent` (subagent_type=general-purpose) | nested `task` agent, await output     | spawn nested CLI session                 | Do the work in-thread (more context cost)          |\n| Ask the user a question     | `AskUserQuestion`                    | inline prompt                            | inline prompt                            | Ask in the response and wait for the next turn     |\n| Subagent model tier         | Tier A: `opus` + `ultrathink`; Tier B: `sonnet` + medium thinking | Tier A: top `task` model + high reasoning; Tier B: smaller-tier `task` + medium reasoning | Tier A: top GPT + `reasoning_effort=high`; Tier B: current spark\u002Fcost-optimized Codex + `reasoning_effort=medium` | Honor the per-phase tier mapping in AGENTS.md; retry the same subagent UP on unavailability |\n\u003C!-- autospec-block:harness-adapter-core -->\n\n**Persistent project notes**: write durable preferences to **`AGENTS.md`** in the repo root — this is the de-facto standard recognized by Claude Code (also reads `CLAUDE.md`), OpenCode, and Codex. Per AGENTS.md, subagent dispatches use a **two-tier policy**: Tier A (top model + extended thinking) for spec work, Tier B (cheaper model + medium thinking) for implementation work. The orchestrator keeps the user's invoked model. Fall back UP the tier on quota\u002Fcapacity or other unavailability by retrying the same subagent with the stronger tier while preserving parent context.\n\n## Harness detection (run once at skill start, before any subcommand)\n\nDetect your harness by checking available tools before any subcommand:\n\n1. **Claude Code** — the `Agent` tool with a `subagent_type` parameter is available.\n   - `TIER_A` = `opus` + `ultrathink`  (the alias, which resolves to the current Claude Opus generation)\n   - `TIER_B` = `sonnet`               (the alias, which resolves to the current Claude Sonnet generation)\n\n2. **OpenCode** — a `task` tool with model\u002Ftier configuration is available (no `subagent_type`).\n   - `TIER_A` = top-tier task model + high reasoning\n   - `TIER_B` = smaller-tier task model + medium reasoning\n\n3. **Codex CLI** — neither `Agent` nor a configurable `task` tool is available; `apply_patch` is the primary edit tool.\n   - `TIER_A` = current top GPT model + `reasoning_effort=high`\n   - `TIER_B` = the spark \u002F cost-optimized variant of the configured model when one exists, else the configured model + `reasoning_effort=medium`\n\n**Fallback rule:** If `TIER_B` is not available in your harness (model unknown, quota\u002Fcapacity failure, authorization failure, or tool call returns an error for that model), silently retry the same subagent dispatch with `TIER_A`. Preserve the parent context on retry; for Codex native subagents, fork\u002Finherit the current conversation context and use the latest top GPT model instead of moving the work into the main session. Never ask the user.\n\nHold `TIER_A` and `TIER_B` for the entire skill run. Every \"Tier A\" and \"Tier B\" reference below resolves to these harness-specific values.\n\n## Catalog source (single source of truth)\n\nFetched at runtime, never vendored:\n\n```bash\nAUTOSPEC_DESIGN_CATALOG_OWNER=\"${AUTOSPEC_DESIGN_CATALOG_OWNER:-berlinguyinca}\"\nAUTOSPEC_DESIGN_CATALOG_REPO=\"${AUTOSPEC_DESIGN_CATALOG_REPO:-awesome-design-md}\"\nAUTOSPEC_DESIGN_CATALOG_REF=\"${AUTOSPEC_DESIGN_CATALOG_REF:-main}\"\n```\n\nPer-vendor DESIGN.md is fetched via `gh api repos\u002F${owner}\u002F${repo}\u002Fcontents\u002Fdesign-md\u002F${vendor}\u002FDESIGN.md?ref=${ref} --jq '.content' | base64 -d`, with a fallback to `curl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002F...`. Cache lands under `~\u002F.autospec\u002Fdesign-cache\u002F\u003Cvendor>\u002FDESIGN.md` with a 24h freshness window. The catalog fetcher script lands in issue #576 (`fetch-design-md.sh`).\n\n## Suggest\n\n> **Model tier:** `TIER_B` (implementation work) — scoring is mechanical; the\n> rubric is a fixed table and the helper is deterministic.\n\n`suggest` scans the current repo, scores every catalog vendor against a fixed\nrubric, and presents the top 3 candidates with a one-line rationale each. It is\n**read-only** — it never modifies any file in the target repo.\n\nRun the deterministic scorer (it prints tab-separated `\u003Cscore>\\t\u003Cvendor>\\t\u003Crationale>`\nlines, highest score first):\n\n```bash\nbash skills\u002Fautospec-design\u002Fscripts\u002Fscore-suggestion.sh \u003Crepo-root>\n```\n\nThe scorer:\n\n1. Fetches the catalog vendor list once (via `gh api ...\u002Fcontents\u002Fdesign-md`,\n   `curl` fallback). When you have already fetched it, pass it in via\n   `AUTOSPEC_DESIGN_VENDORS` to avoid a second round-trip.\n2. Detects repo signals: framework (`package.json`, `next.config.*`,\n   `vite.config.*`, `angular.json`, `svelte.config.*`, Tailwind config), brand\n   keywords (README + `package.json` `name`\u002F`description`), and product-domain\n   keywords (README).\n3. Scores each vendor on the rubric below (cap 6):\n   - **Framework match (+2)** — repo uses a web\u002FJS UI framework AND the vendor\n     is a developer \u002F SaaS \u002F web-product brand (e.g. Linear, Vercel, Stripe).\n   - **Brand match (+1)** — the vendor's normalized name appears in the repo's\n     README or `package.json`.\n   - **Domain match (+1)** — a vendor-name token overlaps the repo README.\n\nPresent the top 3 to the user with their rationales and let them pick. Then run\n`\u002Fautospec-design apply \u003Cvendor>` with the chosen vendor. Do **not** apply\nautomatically — `suggest` only recommends.\n\n## Apply\n\n> **Model tier:** `TIER_B` (implementation work) — write-out is a deterministic\n> branch-and-commit; no novel design.\n\n`apply \u003Cvendor> [--force] [--branch \u003Cname>]` fetches `\u003Cvendor>\u002FDESIGN.md` from\nthe catalog and writes it to `\u003Crepo-root>\u002FDESIGN.md` on a fresh feature branch,\ncommits, and pushes. It is **mechanical** — no subagent dispatch. The flow:\n\n1. Parse `\u003Cvendor>` plus optional `--force` and `--branch \u003Cname>`.\n2. Validate `\u003Cvendor>` exists in the catalog by delegating to\n   `fetch-design-md.sh` (its exit code is the validation — a missing vendor\n   exits non-zero and prints Levenshtein hints; do not duplicate that logic).\n3. If `DESIGN.md` already exists at the repo root and `--force` is absent:\n   refuse, print a `git diff` hint, and exit non-zero. Never silently\n   overwrite.\n4. Refuse to commit on `main` — when `--branch` is absent, auto-create\n   `feat\u002Fdesign-\u003Cvendor>`; otherwise use the supplied branch name.\n5. Fetch the DESIGN.md body (cache hit or network) via `fetch-design-md.sh`.\n6. Write the body to `\u003Crepo-root>\u002FDESIGN.md`.\n7. `git add DESIGN.md`, commit with `feat(design): adopt \u003Cvendor> design\n   language`, and push the branch (`git push -u origin \u003Cbranch>`).\n8. Print a status summary: branch name, DESIGN.md byte count, and the\n   next-step hint (`open a PR, or run \u002Fautospec-design migrate \u003Cvendor>`).\n\n`apply` never opens a PR — the operator does that, or `\u002Fautospec-design\nmigrate` takes over. It is idempotent: a second run on the same branch with the\nsame vendor rewrites identical bytes and exits without creating a duplicate\ncommit.\n\nThe deterministic flow below is the canonical script for the steps above. Run\nit verbatim (it reads `\u003Cvendor>` from the invocation; the `DESIGN_*` overrides\nexist only so the test harness can point it at a fixture repo and a seeded\ncache). It honors `--force`\u002F`--branch` via the `DESIGN_FORCE`\u002F`DESIGN_BRANCH`\nmappings the wrapper sets from the parsed flags.\n\n```bash\n# autospec-design-apply-flow\nset -u\nvendor=\"${DESIGN_VENDOR:?vendor required: \u002Fautospec-design apply \u003Cvendor>}\"\nrepo_root=\"${DESIGN_REPO_ROOT:-$(git rev-parse --show-toplevel)}\"\nfetch=\"${DESIGN_FETCH:-skills\u002Fautospec-design\u002Fscripts\u002Ffetch-design-md.sh}\"\nbranch=\"${DESIGN_BRANCH:-feat\u002Fdesign-$vendor}\"\ntarget=\"$repo_root\u002FDESIGN.md\"\n\n# Guard: refuse overwrite without --force.\nif [ -e \"$target\" ] && [ \"${DESIGN_FORCE:-0}\" != \"1\" ]; then\n    printf 'autospec-design apply: %s already exists. Re-run with --force to overwrite (inspect with: git -C %q diff -- DESIGN.md).\\n' \\\n        \"$target\" \"$repo_root\" >&2\n    exit 1\nfi\n\n# Validate vendor + fetch body (non-zero here = unknown vendor or network).\nbody=\"$(bash \"$fetch\" \"$vendor\")\" || exit $?\n\n# Never commit on main: create\u002Fswitch to the feature branch.\ngit -C \"$repo_root\" checkout -B \"$branch\" >\u002Fdev\u002Fnull 2>&1 || {\n    printf 'autospec-design apply: could not create branch %s\\n' \"$branch\" >&2\n    exit 1\n}\n\n# Write, stage, commit.\nprintf '%s' \"$body\" > \"$target\"\ngit -C \"$repo_root\" add DESIGN.md\nif git -C \"$repo_root\" diff --cached --quiet -- DESIGN.md; then\n    printf 'autospec-design apply: DESIGN.md already matches %s on branch %s.\\n' \\\n        \"$vendor\" \"$branch\"\nelse\n    git -C \"$repo_root\" commit -q -m \"feat(design): adopt $vendor design language\"\nfi\n\n# Push unless the test harness suppresses it.\nif [ \"${DESIGN_NO_PUSH:-0}\" != \"1\" ]; then\n    git -C \"$repo_root\" push -u origin \"$branch\"\nfi\n\nbytes=\"$(wc -c \u003C \"$target\" | tr -d ' ')\"\nprintf 'autospec-design apply: wrote %s (%s bytes) on branch %s.\\n' \\\n    \"$target\" \"$bytes\" \"$branch\"\nprintf 'Next: open a PR, or run \u002Fautospec-design migrate %s.\\n' \"$vendor\"\n```\n\n## Migrate\n\n> **Model tier:** `TIER_B` (implementation work) — emits a structured spec and\n> hands off to `\u002Fautospec-define`.\n\n`migrate \u003Cvendor> [--branch \u003Cname>]` turns an applied vendor `DESIGN.md` into a\ntracked migration spec and then hands that spec to `\u002Fautospec-define` for normal\nautospec issue decomposition. It assumes `\u002Fautospec-design apply \u003Cvendor>` has\nalready written the repo-root `DESIGN.md`.\n\nPreconditions and flow:\n\n1. Verify `\u003Crepo-root>\u002FDESIGN.md` exists and its first heading or first 20 lines\n   mention `\u003Cvendor>` (case-insensitive, with suffixes like `.app` ignored).\n   Otherwise print `Run \u002Fautospec-design apply \u003Cvendor> first.` and exit\n   non-zero.\n2. Run `scan-ui-sources.sh \u003Crepo-root>`; if no UI files are found, surface the\n   helper's no-UI-files message and exit non-zero.\n3. Run `gen-migration-spec.sh \u003Cvendor> \u003Crepo-root>` to write\n   `docs\u002Fspecs\u002F\u003CYYYY-MM-DD>-design-migration-\u003Cvendor>.md`.\n4. Commit the spec on the current feature branch (or create\u002Fswitch to\n   `--branch \u003Cname>` first when provided).\n5. Hand off to `\u002Fautospec-define \u003Cspec-path>`. If that handoff fails, do not\n   swallow the error: leave the committed spec on disk and report the exact\n   command for manual recovery.\n\nThe deterministic flow below is the canonical shell portion for steps 1-4. The\nactual harness handoff is the final printed `\u002Fautospec-define \u003Cspec-path>` line;\ninvoke that skill through the harness after this script succeeds.\n\n```bash\n# autospec-design-migrate-flow\nset -u\nvendor=\"${DESIGN_VENDOR:?vendor required: \u002Fautospec-design migrate \u003Cvendor>}\"\nrepo_root=\"${DESIGN_REPO_ROOT:-$(git rev-parse --show-toplevel)}\"\nbranch=\"${DESIGN_BRANCH:-}\"\nscan=\"${DESIGN_SCAN:-skills\u002Fautospec-design\u002Fscripts\u002Fscan-ui-sources.sh}\"\ngenerate=\"${DESIGN_GENERATE:-skills\u002Fautospec-design\u002Fscripts\u002Fgen-migration-spec.sh}\"\ntarget=\"$repo_root\u002FDESIGN.md\"\nvendor_key=\"$(printf '%s' \"$vendor\" | sed -E 's\u002F[.](app|ai|com|io)$\u002F\u002F' | tr '[:upper:]' '[:lower:]')\"\n\nif [ ! -f \"$target\" ] || ! head -20 \"$target\" | tr '[:upper:]' '[:lower:]' | grep -qF \"$vendor_key\"; then\n    printf 'Run \u002Fautospec-design apply %s first.\\n' \"$vendor\" >&2\n    exit 1\nfi\n\nif [ -n \"$branch\" ]; then\n    git -C \"$repo_root\" checkout -B \"$branch\" >\u002Fdev\u002Fnull 2>&1 || {\n        printf 'autospec-design migrate: could not create branch %s\\n' \"$branch\" >&2\n        exit 1\n    }\nfi\n\n# Prove the UI inventory is non-empty before generating the spec.\ninventory=\"$(bash \"$scan\" \"$repo_root\")\" || exit $?\nif printf '%s' \"$inventory\" | grep -q '\"files\":\\[\\]'; then\n    printf 'gen-migration-spec: no UI files found in %s; nothing to migrate.\\n' \"$repo_root\" >&2\n    exit 2\nfi\n\nspec_path=\"$(bash \"$generate\" \"$vendor\" \"$repo_root\")\" || exit $?\ngit -C \"$repo_root\" add \"$spec_path\"\nif git -C \"$repo_root\" diff --cached --quiet -- \"$spec_path\"; then\n    printf 'autospec-design migrate: migration spec already committed: %s\\n' \"$spec_path\"\nelse\n    git -C \"$repo_root\" commit -q -m \"docs(design): add $vendor migration spec\"\nfi\n\nprintf 'autospec-design migrate: wrote %s.\\n' \"$spec_path\"\nprintf '\u002Fautospec-define %s\\n' \"$spec_path\"\n```\n\n## Hard rules\n\n- Never overwrite an existing `DESIGN.md` without `--force`.\n- Never open a PR automatically — the operator opens it (or `\u002Fautospec-define`\n  takes over via the migration spec).\n- Never modify the issue title (when running via the autospec pipeline).\n- Always idempotent — running `apply` twice with the same vendor on the same\n  branch is a no-op after the first commit.\n- `gh` CLI only for GitHub ops; `curl` only as a fallback for catalog reads.\n","# autospec-design (harness-neutral)\n\nSkill scaffold for the **autospec-design** workflow. The full workflow lets the\nuser pick a vendor design language (Linear, Stripe, Vercel, ...) from a curated\ncatalog, write `DESIGN.md` at the project root, and optionally hand off a\nmigration spec into `\u002Fautospec-define`. The catalog lives at\n`berlinguyinca\u002Fawesome-design-md` and is fetched at runtime — never vendored.\n\nThis file is the **autospec-design trio body**. The `suggest` and `apply`\nsubcommands are implemented; `migrate` still holds a placeholder until issue\n#580 lands.\n\nManage your own context — never exceed 60%. Delegate to subagents whenever your harness supports it.\n\n\u003C!-- autospec-block:startup-self-update SKILL_NAME=autospec-design -->\n\n## Self-update mode\n\nIf the feature-request argument matches the regex `^\\s*update\\s*$` (case-insensitive, whitespace-padded), this skill enters self-update mode and does not run the normal pipeline:\n\n1. **Detect harness** by checking which install path exists for this skill:\n   - Claude Code: `~\u002F.claude\u002Fskills\u002Fautospec-design\u002FSKILL.md`\n   - OpenCode:    `~\u002F.config\u002Fopencode\u002Fagent\u002Fautospec-design.md`\n   - Codex CLI:   `~\u002F.codex\u002Fprompts\u002Fautospec-design.md`\n2. **Re-install the full autospec suite from `main`** by piping the canonical installer:\n   ```bash\n   curl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002Fberlinguyinca\u002Fautospec\u002Fmain\u002Fbootstrap.sh | bash -s -- --skill all --harness all --update\n   ```\n   Run this one-liner once; it refreshes all autospec skills across all harnesses.\n3. **Show the diff** between the prior installed file(s) and the freshly fetched copy (e.g. `diff \u003C(cat \u003Cprior>) \u003C(curl -fsSL ...SKILL.md)` or the equivalent recorded by the installer).\n4. **Stop.** Do not enter any subcommand. Print the upgrade summary and return to the user.\n\nIf no install path is detected, print `Self-update: no installed copy of autospec-design found; run install.sh first.` and exit.\n\n## Invocation\n\n```\n\u002Fautospec-design suggest\n\u002Fautospec-design apply \u003Cvendor> [--force] [--branch \u003Cname>]\n\u002Fautospec-design migrate \u003Cvendor> [--branch \u003Cname>]\n```\n\n- `suggest` — scan the current repo, score every catalog vendor by the\n  framework \u002F domain \u002F brand-language rubric, present the top 3 with a\n  one-line rationale each. Prints only — never modifies any files.\n- `apply \u003Cvendor>` — fetch `\u003Cvendor>\u002FDESIGN.md` from the catalog, write it\n  to `\u003Crepo-root>\u002FDESIGN.md` on a fresh feature branch (default:\n  `feat\u002Fdesign-\u003Cvendor>`), commit, and push. Refuses to overwrite an\n  existing `DESIGN.md` without `--force`. Does **not** open a PR.\n- `migrate \u003Cvendor>` — generate a migration spec at\n  `docs\u002Fspecs\u002F\u003CYYYY-MM-DD>-design-migration-\u003Cvendor>.md` from the scanned\n  UI inventory and hand off to `\u002Fautospec-define \u003Cspec-path>`.\n\nThe skill exits non-zero if `gh auth status` fails when issuing a network\noperation that needs authentication. See spec\n`docs\u002Fspecs\u002F2026-05-26-autospec-design-skill.md` for the full contract.\n\n## Required capabilities & harness adapter\n\nThis workflow assumes a small set of capabilities. Map each one to your harness's actual tool. If a capability is missing, use the listed fallback.\n\n| Capability                  | Claude Code                          | OpenCode                                 | Codex CLI                                | Fallback if missing                                |\n|-----------------------------|--------------------------------------|------------------------------------------|------------------------------------------|----------------------------------------------------|\n| Read-only codebase research | `Agent` (subagent_type=Explore)      | `task` agent in read-only mode           | `apply_patch` read-only \u002F shell `grep`   | Do the search in-thread with `rg`\u002F`grep`           |\n| Foreground delegation       | `Agent` (subagent_type=general-purpose) | nested `task` agent, await output     | spawn nested CLI session                 | Do the work in-thread (more context cost)          |\n| Ask the user a question     | `AskUserQuestion`                    | inline prompt                            | inline prompt                            | Ask in the response and wait for the next turn     |\n| Subagent model tier         | Tier A: `opus` + `ultrathink`; Tier B: `sonnet` + medium thinking | Tier A: top `task` model + high reasoning; Tier B: smaller-tier `task` + medium reasoning | Tier A: top GPT + `reasoning_effort=high`; Tier B: current spark\u002Fcost-optimized Codex + `reasoning_effort=medium` | Honor the per-phase tier mapping in AGENTS.md; retry the same subagent UP on unavailability |\n\u003C!-- autospec-block:harness-adapter-core -->\n\n**Persistent project notes**: write durable preferences to **`AGENTS.md`** in the repo root — this is the de-facto standard recognized by Claude Code (also reads `CLAUDE.md`), OpenCode, and Codex. Per AGENTS.md, subagent dispatches use a **two-tier policy**: Tier A (top model + extended thinking) for spec work, Tier B (cheaper model + medium thinking) for implementation work. The orchestrator keeps the user's invoked model. Fall back UP the tier on quota\u002Fcapacity or other unavailability by retrying the same subagent with the stronger tier while preserving parent context.\n\n## Harness detection (run once at skill start, before any subcommand)\n\nDetect your harness by checking available tools before any subcommand:\n\n1. **Claude Code** — the `Agent` tool with a `subagent_type` parameter is available.\n   - `TIER_A` = `opus` + `ultrathink`  (the alias, which resolves to the current Claude Opus generation)\n   - `TIER_B` = `sonnet`               (the alias, which resolves to the current Claude Sonnet generation)\n\n2. **OpenCode** — a `task` tool with model\u002Ftier configuration is available (no `subagent_type`).\n   - `TIER_A` = top-tier task model + high reasoning\n   - `TIER_B` = smaller-tier task model + medium reasoning\n\n3. **Codex CLI** — neither `Agent` nor a configurable `task` tool is available; `apply_patch` is the primary edit tool.\n   - `TIER_A` = current top GPT model + `reasoning_effort=high`\n   - `TIER_B` = the spark \u002F cost-optimized variant of the configured model when one exists, else the configured model + `reasoning_effort=medium`\n\n**Fallback rule:** If `TIER_B` is not available in your harness (model unknown, quota\u002Fcapacity failure, authorization failure, or tool call returns an error for that model), silently retry the same subagent dispatch with `TIER_A`. Preserve the parent context on retry; for Codex native subagents, fork\u002Finherit the current conversation context and use the latest top GPT model instead of moving the work into the main session. Never ask the user.\n\nHold `TIER_A` and `TIER_B` for the entire skill run. Every \"Tier A\" and \"Tier B\" reference below resolves to these harness-specific values.\n\n## Catalog source (single source of truth)\n\nFetched at runtime, never vendored:\n\n```bash\nAUTOSPEC_DESIGN_CATALOG_OWNER=\"${AUTOSPEC_DESIGN_CATALOG_OWNER:-berlinguyinca}\"\nAUTOSPEC_DESIGN_CATALOG_REPO=\"${AUTOSPEC_DESIGN_CATALOG_REPO:-awesome-design-md}\"\nAUTOSPEC_DESIGN_CATALOG_REF=\"${AUTOSPEC_DESIGN_CATALOG_REF:-main}\"\n```\n\nPer-vendor DESIGN.md is fetched via `gh api repos\u002F${owner}\u002F${repo}\u002Fcontents\u002Fdesign-md\u002F${vendor}\u002FDESIGN.md?ref=${ref} --jq '.content' | base64 -d`, with a fallback to `curl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002F...`. Cache lands under `~\u002F.autospec\u002Fdesign-cache\u002F\u003Cvendor>\u002FDESIGN.md` with a 24h freshness window. The catalog fetcher script lands in issue #576 (`fetch-design-md.sh`).\n\n## Suggest\n\n> **Model tier:** `TIER_B` (implementation work) — scoring is mechanical; the\n> rubric is a fixed table and the helper is deterministic.\n\n`suggest` scans the current repo, scores every catalog vendor against a fixed\nrubric, and presents the top 3 candidates with a one-line rationale each. It is\n**read-only** — it never modifies any file in the target repo.\n\nRun the deterministic scorer (it prints tab-separated `\u003Cscore>\\t\u003Cvendor>\\t\u003Crationale>`\nlines, highest score first):\n\n```bash\nbash skills\u002Fautospec-design\u002Fscripts\u002Fscore-suggestion.sh \u003Crepo-root>\n```\n\nThe scorer:\n\n1. Fetches the catalog vendor list once (via `gh api ...\u002Fcontents\u002Fdesign-md`,\n   `curl` fallback). When you have already fetched it, pass it in via\n   `AUTOSPEC_DESIGN_VENDORS` to avoid a second round-trip.\n2. Detects repo signals: framework (`package.json`, `next.config.*`,\n   `vite.config.*`, `angular.json`, `svelte.config.*`, Tailwind config), brand\n   keywords (README + `package.json` `name`\u002F`description`), and product-domain\n   keywords (README).\n3. Scores each vendor on the rubric below (cap 6):\n   - **Framework match (+2)** — repo uses a web\u002FJS UI framework AND the vendor\n     is a developer \u002F SaaS \u002F web-product brand (e.g. Linear, Vercel, Stripe).\n   - **Brand match (+1)** — the vendor's normalized name appears in the repo's\n     README or `package.json`.\n   - **Domain match (+1)** — a vendor-name token overlaps the repo README.\n\nPresent the top 3 to the user with their rationales and let them pick. Then run\n`\u002Fautospec-design apply \u003Cvendor>` with the chosen vendor. Do **not** apply\nautomatically — `suggest` only recommends.\n\n## Apply\n\n> **Model tier:** `TIER_B` (implementation work) — write-out is a deterministic\n> branch-and-commit; no novel design.\n\n`apply \u003Cvendor> [--force] [--branch \u003Cname>]` fetches `\u003Cvendor>\u002FDESIGN.md` from\nthe catalog and writes it to `\u003Crepo-root>\u002FDESIGN.md` on a fresh feature branch,\ncommits, and pushes. It is **mechanical** — no subagent dispatch. The flow:\n\n1. Parse `\u003Cvendor>` plus optional `--force` and `--branch \u003Cname>`.\n2. Validate `\u003Cvendor>` exists in the catalog by delegating to\n   `fetch-design-md.sh` (its exit code is the validation — a missing vendor\n   exits non-zero and prints Levenshtein hints; do not duplicate that logic).\n3. If `DESIGN.md` already exists at the repo root and `--force` is absent:\n   refuse, print a `git diff` hint, and exit non-zero. Never silently\n   overwrite.\n4. Refuse to commit on `main` — when `--branch` is absent, auto-create\n   `feat\u002Fdesign-\u003Cvendor>`; otherwise use the supplied branch name.\n5. Fetch the DESIGN.md body (cache hit or network) via `fetch-design-md.sh`.\n6. Write the body to `\u003Crepo-root>\u002FDESIGN.md`.\n7. `git add DESIGN.md`, commit with `feat(design): adopt \u003Cvendor> design\n   language`, and push the branch (`git push -u origin \u003Cbranch>`).\n8. Print a status summary: branch name, DESIGN.md byte count, and the\n   next-step hint (`open a PR, or run \u002Fautospec-design migrate \u003Cvendor>`).\n\n`apply` never opens a PR — the operator does that, or `\u002Fautospec-design\nmigrate` takes over. It is idempotent: a second run on the same branch with the\nsame vendor rewrites identical bytes and exits without creating a duplicate\ncommit.\n\nThe deterministic flow below is the canonical script for the steps above. Run\nit verbatim (it reads `\u003Cvendor>` from the invocation; the `DESIGN_*` overrides\nexist only so the test harness can point it at a fixture repo and a seeded\ncache). It honors `--force`\u002F`--branch` via the `DESIGN_FORCE`\u002F`DESIGN_BRANCH`\nmappings the wrapper sets from the parsed flags.\n\n```bash\n# autospec-design-apply-flow\nset -u\nvendor=\"${DESIGN_VENDOR:?vendor required: \u002Fautospec-design apply \u003Cvendor>}\"\nrepo_root=\"${DESIGN_REPO_ROOT:-$(git rev-parse --show-toplevel)}\"\nfetch=\"${DESIGN_FETCH:-skills\u002Fautospec-design\u002Fscripts\u002Ffetch-design-md.sh}\"\nbranch=\"${DESIGN_BRANCH:-feat\u002Fdesign-$vendor}\"\ntarget=\"$repo_root\u002FDESIGN.md\"\n\n# Guard: refuse overwrite without --force.\nif [ -e \"$target\" ] && [ \"${DESIGN_FORCE:-0}\" != \"1\" ]; then\n    printf 'autospec-design apply: %s already exists. Re-run with --force to overwrite (inspect with: git -C %q diff -- DESIGN.md).\\n' \\\n        \"$target\" \"$repo_root\" >&2\n    exit 1\nfi\n\n# Validate vendor + fetch body (non-zero here = unknown vendor or network).\nbody=\"$(bash \"$fetch\" \"$vendor\")\" || exit $?\n\n# Never commit on main: create\u002Fswitch to the feature branch.\ngit -C \"$repo_root\" checkout -B \"$branch\" >\u002Fdev\u002Fnull 2>&1 || {\n    printf 'autospec-design apply: could not create branch %s\\n' \"$branch\" >&2\n    exit 1\n}\n\n# Write, stage, commit.\nprintf '%s' \"$body\" > \"$target\"\ngit -C \"$repo_root\" add DESIGN.md\nif git -C \"$repo_root\" diff --cached --quiet -- DESIGN.md; then\n    printf 'autospec-design apply: DESIGN.md already matches %s on branch %s.\\n' \\\n        \"$vendor\" \"$branch\"\nelse\n    git -C \"$repo_root\" commit -q -m \"feat(design): adopt $vendor design language\"\nfi\n\n# Push unless the test harness suppresses it.\nif [ \"${DESIGN_NO_PUSH:-0}\" != \"1\" ]; then\n    git -C \"$repo_root\" push -u origin \"$branch\"\nfi\n\nbytes=\"$(wc -c \u003C \"$target\" | tr -d ' ')\"\nprintf 'autospec-design apply: wrote %s (%s bytes) on branch %s.\\n' \\\n    \"$target\" \"$bytes\" \"$branch\"\nprintf 'Next: open a PR, or run \u002Fautospec-design migrate %s.\\n' \"$vendor\"\n```\n\n## Migrate\n\n> **Model tier:** `TIER_B` (implementation work) — emits a structured spec and\n> hands off to `\u002Fautospec-define`.\n\n`migrate \u003Cvendor> [--branch \u003Cname>]` turns an applied vendor `DESIGN.md` into a\ntracked migration spec and then hands that spec to `\u002Fautospec-define` for normal\nautospec issue decomposition. It assumes `\u002Fautospec-design apply \u003Cvendor>` has\nalready written the repo-root `DESIGN.md`.\n\nPreconditions and flow:\n\n1. Verify `\u003Crepo-root>\u002FDESIGN.md` exists and its first heading or first 20 lines\n   mention `\u003Cvendor>` (case-insensitive, with suffixes like `.app` ignored).\n   Otherwise print `Run \u002Fautospec-design apply \u003Cvendor> first.` and exit\n   non-zero.\n2. Run `scan-ui-sources.sh \u003Crepo-root>`; if no UI files are found, surface the\n   helper's no-UI-files message and exit non-zero.\n3. Run `gen-migration-spec.sh \u003Cvendor> \u003Crepo-root>` to write\n   `docs\u002Fspecs\u002F\u003CYYYY-MM-DD>-design-migration-\u003Cvendor>.md`.\n4. Commit the spec on the current feature branch (or create\u002Fswitch to\n   `--branch \u003Cname>` first when provided).\n5. Hand off to `\u002Fautospec-define \u003Cspec-path>`. If that handoff fails, do not\n   swallow the error: leave the committed spec on disk and report the exact\n   command for manual recovery.\n\nThe deterministic flow below is the canonical shell portion for steps 1-4. The\nactual harness handoff is the final printed `\u002Fautospec-define \u003Cspec-path>` line;\ninvoke that skill through the harness after this script succeeds.\n\n```bash\n# autospec-design-migrate-flow\nset -u\nvendor=\"${DESIGN_VENDOR:?vendor required: \u002Fautospec-design migrate \u003Cvendor>}\"\nrepo_root=\"${DESIGN_REPO_ROOT:-$(git rev-parse --show-toplevel)}\"\nbranch=\"${DESIGN_BRANCH:-}\"\nscan=\"${DESIGN_SCAN:-skills\u002Fautospec-design\u002Fscripts\u002Fscan-ui-sources.sh}\"\ngenerate=\"${DESIGN_GENERATE:-skills\u002Fautospec-design\u002Fscripts\u002Fgen-migration-spec.sh}\"\ntarget=\"$repo_root\u002FDESIGN.md\"\nvendor_key=\"$(printf '%s' \"$vendor\" | sed -E 's\u002F[.](app|ai|com|io)$\u002F\u002F' | tr '[:upper:]' '[:lower:]')\"\n\nif [ ! -f \"$target\" ] || ! head -20 \"$target\" | tr '[:upper:]' '[:lower:]' | grep -qF \"$vendor_key\"; then\n    printf 'Run \u002Fautospec-design apply %s first.\\n' \"$vendor\" >&2\n    exit 1\nfi\n\nif [ -n \"$branch\" ]; then\n    git -C \"$repo_root\" checkout -B \"$branch\" >\u002Fdev\u002Fnull 2>&1 || {\n        printf 'autospec-design migrate: could not create branch %s\\n' \"$branch\" >&2\n        exit 1\n    }\nfi\n\n# Prove the UI inventory is non-empty before generating the spec.\ninventory=\"$(bash \"$scan\" \"$repo_root\")\" || exit $?\nif printf '%s' \"$inventory\" | grep -q '\"files\":\\[\\]'; then\n    printf 'gen-migration-spec: no UI files found in %s; nothing to migrate.\\n' \"$repo_root\" >&2\n    exit 2\nfi\n\nspec_path=\"$(bash \"$generate\" \"$vendor\" \"$repo_root\")\" || exit $?\ngit -C \"$repo_root\" add \"$spec_path\"\nif git -C \"$repo_root\" diff --cached --quiet -- \"$spec_path\"; then\n    printf 'autospec-design migrate: migration spec already committed: %s\\n' \"$spec_path\"\nelse\n    git -C \"$repo_root\" commit -q -m \"docs(design): add $vendor migration spec\"\nfi\n\nprintf 'autospec-design migrate: wrote %s.\\n' \"$spec_path\"\nprintf '\u002Fautospec-define %s\\n' \"$spec_path\"\n```\n\n## Hard rules\n\n- Never overwrite an existing `DESIGN.md` without `--force`.\n- Never open a PR automatically — the operator opens it (or `\u002Fautospec-define`\n  takes over via the migration spec).\n- Never modify the issue title (when running via the autospec pipeline).\n- Always idempotent — running `apply` twice with the same vendor on the same\n  branch is a no-op after the first commit.\n- `gh` CLI only for GitHub ops; `curl` only as a fallback for catalog reads.\n",{"name":5,"description":7},"main","9e7f04357ff46ab731d4f56a4ac5ca5449c13bb0","skills\u002Fautospec-design",1,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec","https:\u002F\u002Fgithub.com\u002Fberlinguyinca","Love to lead, code, cook and contribute to open source and believe that you can learn anything you set your mind too.\r\n","2026-11-11 07:09:35",{"source":8,"skillName":5,"commands":63,"prompt":73,"zipAvailable":74},[64,67,70],{"tool":65,"command":66},"npx","npx skills add berlinguyinca\u002Fautospec --skill autospec-design",{"tool":68,"command":69},"bunx","bunx skills add berlinguyinca\u002Fautospec --skill autospec-design",{"tool":71,"command":72},"pnpm","pnpm dlx skills add berlinguyinca\u002Fautospec --skill autospec-design","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002FSKILL.md",true,[76,82,88,92,96,100,104,108,112],{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":80,"sourceHtmlUrl":81},"codex\u002Fprompt.md","REFERENCE","参考文档",16405,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Fcodex\u002Fprompt.md",{"path":83,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":86,"sourceHtmlUrl":87},"install.sh","SCRIPT","脚本",10798,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Finstall.sh",{"path":89,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":90,"sourceHtmlUrl":91},"opencode\u002Fagent.md",16791,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Fopencode\u002Fagent.md",{"path":93,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":94,"sourceHtmlUrl":95},"README.md",3245,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002FREADME.md",{"path":97,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":98,"sourceHtmlUrl":99},"scripts\u002Ffetch-design-md.sh",7426,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Fscripts\u002Ffetch-design-md.sh",{"path":101,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":102,"sourceHtmlUrl":103},"scripts\u002Fgen-migration-spec.sh",3126,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Fscripts\u002Fgen-migration-spec.sh",{"path":105,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":106,"sourceHtmlUrl":107},"scripts\u002Fscan-ui-sources.sh",2195,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Fscripts\u002Fscan-ui-sources.sh",{"path":109,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":110,"sourceHtmlUrl":111},"scripts\u002Fscore-suggestion.sh",7442,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Fscripts\u002Fscore-suggestion.sh",{"path":113,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":114,"sourceHtmlUrl":115},"uninstall.sh",3947,"https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-design\u002Funinstall.sh",[117,128,137,146,168,190],{"skillNo":118,"name":119,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":125,"publishedAt":49},"01M71RJHYMKPTC9AAPXWQ6F4RT","trio-pass-skill","https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Flockstep-trio-pass\u002FSKILL.md",[122],{"categoryNo":123,"categoryName":124,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[126,127],"trio","trio pass skill",{"skillNo":129,"name":130,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":131,"categories":132,"keywords":134,"publishedAt":49},"01M71RJFBQ1KAF9XM3T3Y4VMSK","duo-pass-skill","https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Flockstep-duo-pass\u002FSKILL.md",[133],{"categoryNo":123,"categoryName":124,"taxonomyType":21,"taxonomyTypeDesc":22},[135,136],"duo pass skill","duo",{"skillNo":138,"name":139,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":140,"categories":141,"keywords":143,"publishedAt":49},"01M71RJDT1TB3P9TRWR6A534FB","duo-divergence-skill","https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Flockstep-duo-divergence\u002FSKILL.md",[142],{"categoryNo":123,"categoryName":124,"taxonomyType":21,"taxonomyTypeDesc":22},[144,136,145],"duo divergence skill","divergence",{"skillNo":147,"name":29,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":149,"categories":150,"keywords":152,"publishedAt":49},"01M71RJAC4NNEQSYBDSY3NZS73","Use when the user wants to ship a feature end-to-end across multiple commits — bootstraps repo if missing, brainstorms a design spec, decomposes into linked GitHub issues, or splits an existing spec into issues, then runs an autonomous implementation loop with auto-merge.","https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec\u002FSKILL.md",[151],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[29,153,154,155,31,156,157,158,42,159,160,161,162,163,164,165,166,46,167,43],"ship a feature","end","issues","autonomous","auto","loop","across","design spec","merge","ship","implementation loop","github issues","feature","commits","missing",{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":174,"publishedAt":49},"01M71RJ4F9HT8PT26CKNM9XE1M","autospec-upgrade","Use when the user wants to upgrade an Angular, Next.js, or React project to the latest official versions safely — by locking observable behavior before touching versions, upgrading one major at a time with official codemods, and gating completion on mutation score rather than coverage theater.","https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-upgrade\u002FSKILL.md",[],[175,29,176,177,178,179,47,180,181,42,182,183,184,34,43,185,186,187,188,189],"autospec upgrade","upgrade","mutation","coverage","official","time","gating","angular","react","next","Next.js","one","with","behavior","latest",{"skillNo":191,"name":192,"nameZh":6,"description":193,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":194,"categories":195,"keywords":197,"publishedAt":49},"01M71RJ28FJ4PM6DXHB7T5CF8H","autospec-ui-audit","Use when a React Router application needs a deterministic whole-site route inventory before UI, UX, navigation, or evidence auditing begins.","https:\u002F\u002Fgithub.com\u002Fberlinguyinca\u002Fautospec\u002Fblob\u002Fmain\u002Fskills\u002Fautospec-ui-audit\u002FSKILL.md",[196],{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},[198,29,199,200,201,43,202,203,204,205,206,207,208,209,210,183,211,42,212],"ui audit","audit","autospec ui audit","ui","inventory","evidence","deterministic","route","ux","site","react router","navigation","application","needs","router"]