From 1a4a25a540ce282dd330b4cf0a457c7740136b6c Mon Sep 17 00:00:00 2001 From: Parfii-bot Date: Tue, 21 Apr 2026 13:38:03 +0800 Subject: [PATCH] refactor(skills): update agent references in 2 skills to kei- prefix MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Mechanical rename of all 14 kit-agent references in skills/*/SKILL.md. Pattern: word-boundary match that excludes "-" on both sides, applied longest-first so "ml-researcher" rewrites before "researcher" (avoids the "kei-ml-kei-researcher" double-prefix trap). skills/new-agent/SKILL.md (14 refs): - Phase 3.3 handoff list (code-implementer/critic/validator mandatory, cost-guardian/ml-implementer/ml-researcher/infra-implementer/ security-auditor conditional) - Phase 3.5 role-template example (Hand off ... to code-implementer ...) - Phase 4 manifest-template example (target = "code-implementer" etc.) - Phase 6 report-block example ("Handoffs: code-implementer, critic, ...") - Description-string ref on Phase 1b (cost-guardian mandatory) skills/research/SKILL.md (5 refs): - `critic` renamed to `kei-critic` inside backticks and bold - Teammate-role mentions that happen to share the name - NOTE: hyphenated compound labels like `web-researcher`, `code-explorer`, `{component}-critic`, `meta-critic` are left untouched — they are ad-hoc teammate-role labels, not kit-agent handoffs. The boundary rule (no "-" immediately before the name) preserves them correctly. Other 4 skills (debug-deep, pr-review, refactor, test-gen) have zero kit-agent references — untouched. Co-Authored-By: Claude Opus 4.7 (1M context) --- skills/new-agent/SKILL.md | 28 ++++++++++++++-------------- skills/research/SKILL.md | 10 +++++----- 2 files changed, 19 insertions(+), 19 deletions(-) diff --git a/skills/new-agent/SKILL.md b/skills/new-agent/SKILL.md index 2f3f6db..a401390 100644 --- a/skills/new-agent/SKILL.md +++ b/skills/new-agent/SKILL.md @@ -113,7 +113,7 @@ This call ALWAYS runs. Q7 (scrapers) is independent of Q2/Q3, so every agent ans "options": [ {"label": "None", "description": "No scraping / data extraction"}, {"label": "Free-tier only", "description": "YouTube API v3, GitHub GraphQL, Telegram Telethon, Twitter twscrape — Tier 1"}, - {"label": "Paid tier", "description": "Apify / Bright Data — HIGH GDPR + cost risk, cost-guardian mandatory"} + {"label": "Paid tier", "description": "Apify / Bright Data — HIGH GDPR + cost risk, kei-cost-guardian mandatory"} ] } ] @@ -204,16 +204,16 @@ If any block is missing: ### 3.3 Compute `handoffs` ALWAYS include: -- `code-implementer` — "generic dev work outside this project's stack" -- `critic` — "anti-pattern / Constructor Pattern sweep on diffs >200 LOC" -- `validator` — "fact-check / citation sanity before commit" +- `kei-code-implementer` — "generic dev work outside this project's stack" +- `kei-critic` — "anti-pattern / Constructor Pattern sweep on diffs >200 LOC" +- `kei-validator` — "fact-check / citation sanity before commit" Conditional additions: -- Q3 == Yes → add `cost-guardian` — "paid API run — pricing + cost estimate + dashboard check" -- Q4 == Yes → add `ml-implementer` — "numerical experiment / training run" AND `ml-researcher` — "literature / prior art lookup" -- Q2 != "None yet" → add `infra-implementer` — "node provisioning / deploy / SSH / container ops" -- Q1 is Rust/Swift/Go (any variant) → add `security-auditor` — "crypto / key handling / network / memory review" -- Q7 == "Paid tier" → add `cost-guardian` if not already present (paid scraping = cost risk even when Q3 == No) — "paid scraper run — Apify/Bright Data pricing + cost estimate + dashboard check" +- Q3 == Yes → add `kei-cost-guardian` — "paid API run — pricing + cost estimate + dashboard check" +- Q4 == Yes → add `kei-ml-implementer` — "numerical experiment / training run" AND `kei-ml-researcher` — "literature / prior art lookup" +- Q2 != "None yet" → add `kei-infra-implementer` — "node provisioning / deploy / SSH / container ops" +- Q1 is Rust/Swift/Go (any variant) → add `kei-security-auditor` — "crypto / key handling / network / memory review" +- Q7 == "Paid tier" → add `kei-cost-guardian` if not already present (paid scraping = cost risk even when Q3 == No) — "paid scraper run — Apify/Bright Data pricing + cost estimate + dashboard check" ### 3.4 Compute `references.extra` @@ -234,7 +234,7 @@ Conditional: `role` — 2-3 sentences you compose from the answers. Template: -> You are the {slug} specialist. You own {one-line description}. Stack: {Q1}. Deploy: {Q2 or "local-only"}. {If Q5: "Sensitive IP: no public deploy, double explicit confirmation required."} Hand off generic dev work to `code-implementer`, numerical work to `ml-implementer` (if ML), and infra to `infra-implementer` (if deployed). +> You are the {slug} specialist. You own {one-line description}. Stack: {Q1}. Deploy: {Q2 or "local-only"}. {If Q5: "Sensitive IP: no public deploy, double explicit confirmation required."} Hand off generic dev work to `kei-code-implementer`, numerical work to `kei-ml-implementer` (if ML), and infra to `kei-infra-implementer` (if deployed). `domain_in` — 5-8 lines derived from Q1/Q2 stack/deploy specifics. Use the existing generic manifests in `_manifests/` as shape references. If unsure, @@ -249,7 +249,7 @@ inherits: - If Q5 == Yes: `"Public deployment without double explicit confirmation (\"yes, deploy\" + \"I confirm publication\")"` - If Q6 == Yes: `"Echoing any secret from in chat — reference paths only"` - If Q4 == Yes: `"Running paid training without cost estimate + single-variant verify first"` -- If Q7 == "Paid tier": `"Paid scraper batch >100 items without \`cost-guardian\` pre-run cost estimate"` + `"LinkedIn paid scrape without legal-review sign-off (BGH Germany Nov 2024 GDPR risk)"` +- If Q7 == "Paid tier": `"Paid scraper batch >100 items without \`kei-cost-guardian\` pre-run cost estimate"` + `"LinkedIn paid scrape without legal-review sign-off (BGH Germany Nov 2024 GDPR risk)"` - If Q5 == Yes AND Q7 != "None": `"Scraper output must never contain PII of subjects whose profile references unfiled patent IP (cross-contamination risk — priority-date leak via enrichment)"` - Files >200 LOC without decomposition (Constructor Pattern). - Picking a non-default language without a documented reason. @@ -280,11 +280,11 @@ inherits: lines. Example: ```toml [[handoff]] - target = "code-implementer" + target = "kei-code-implementer" trigger = "generic dev work outside this project's stack" [[handoff]] - target = "critic" + target = "kei-critic" trigger = "anti-pattern / Constructor Pattern sweep on diffs >200 LOC" ``` 5. Write the filled manifest to @@ -334,7 +334,7 @@ Show a concise block to the user: Agent generated: -specialist Blocks: baseline, evidence-grading, memory-protocol, rule-pre-dev-gate, , , - Handoffs: code-implementer, critic, validator, + Handoffs: kei-code-implementer, kei-critic, kei-validator, Manifest: ~/.claude/agents/_manifests/-specialist.toml Generated: ~/.claude/agents/-specialist.md Memory: ~/.claude/memory/-project.md (not yet created — adjust path if your memory layout differs) diff --git a/skills/research/SKILL.md b/skills/research/SKILL.md index 408bf5e..5535e20 100644 --- a/skills/research/SKILL.md +++ b/skills/research/SKILL.md @@ -82,7 +82,7 @@ Route to matching section based on pair {A|B|C, 1|2|3}: Control-level logic applies to ALL variants: - **L1** — before EVERY `TaskCreate` + agent spawn, invoke `AskUserQuestion` with options {Approve | Modify | Skip} - **L2** — auto-spawn discovery waves; `AskUserQuestion` ONLY before verification waves + new-branch spawns -- **L3** — no user prompts; apply `critic` teammate self-evaluation on each wave output before proceeding +- **L3** — no user prompts; apply `kei-critic` teammate self-evaluation on each wave output before proceeding --- @@ -92,7 +92,7 @@ Control-level logic applies to ALL variants: **Phase 1:** Spawn 3 teammates in parallel: - `web-researcher` — WebSearch/WebFetch top 10 findings -- `critic` — limitations, alternatives, failure stories +- `kei-critic` — limitations, alternatives, failure stories - `practical` — real-world use cases + prior art **Phase 2:** Cross-reference + verify + confidence grading (you, lead). @@ -203,7 +203,7 @@ Plus `graph.json` cumulative update + `intersections.md`. > Idle teammates from Phase 1 get reused in Phase 5 — no wasted spawns. > > **Teammate naming convention:** -> - Phase 1: `web-researcher`, `code-explorer`, `critic`, `practical`, `docs` +> - Phase 1: `web-researcher`, `code-explorer`, `kei-critic`, `practical`, `docs` > - Phase 5: reuse same teammates with new tasks (they keep context!) > - If topic doesn't need code exploration, skip `code-explorer` — spawn only what's needed > @@ -241,7 +241,7 @@ Create tasks and assign to teammates: 2. **code-explorer** — If topic relates to code/library, search the codebase and npm/pypi/github for implementations, examples, patterns. [CONDITIONAL: skip if not code-related] -3. **critic** — Search for criticisms, limitations, known issues, alternatives. Find "X vs Y" comparisons, migration guides, deprecation notices. +3. **kei-critic** — Search for criticisms, limitations, known issues, alternatives. Find "X vs Y" comparisons, migration guides, deprecation notices. 4. **practical** — Find real-world usage examples, case studies, production stories. Check GitHub issues, Stack Overflow, blog posts. @@ -311,7 +311,7 @@ Create new tasks and assign to teammates (reuse existing first, spawn additional 1. **web-researcher** (reuse) → **Numbers** — Verify ALL numbers, unit economics, metrics, benchmarks. Recalculate independently. Find primary sources for each figure. If numbers don't agree — mark `[DISPUTED]`. -2. **critic** (reuse) → **Critique** — Devil's advocate. Find ALL reasons this WON'T work. Worst-case scenarios. Legal risks. Ethical problems. What skeptics say. Real failure stories. +2. **kei-critic** (reuse) → **Critique** — Devil's advocate. Find ALL reasons this WON'T work. Worst-case scenarios. Legal risks. Ethical problems. What skeptics say. Real failure stories. 3. **practical** (reuse) → **Competitors** — Find ALL competitors (not only the obvious ones). Check: market map, Crunchbase, ProductHunt, G2/Capterra. Who launched in the last 6 months? Who died? Who pivoted?