F-e-u-e-r
UserDistilled operating skills for daily-driver Claude models — few dense rules, executable gates over long prose.
Categories
Indexed Skills (31)
design-review-gate
Turns design review into observable checks. Load when you are judging a built UI surface ("does this look right/professional?", a visual QA request, before/after comparison of styling work), when generated design variants need comparison, or when any design document is about to be adopted, authored, or treated as authoritative - a DESIGN.md, tokens file, brand guide, or a reverse-engineered competitor analysis. NOT for producing the surface (ui-design-craft, motion-craft), for evidence rules on non-design deliverables (domain-evidence-discipline), or for code-correctness review (the code-review tooling).
motion-craft
Implementation-agnostic motion judgment for product UI. Load when you are adding, reviewing, or tuning animation in any framework - durations, easings, springs, staggers, entrances/exits, gesture handoff, reduced motion - when a spec says "smooth" or "delightful" with no numbers attached, or when you notice animation being judged by feel instead of against a budget. NOT for debugging one animation library's API (use its docs), for static visual decisions (ui-design-craft), or for running a design review end-to-end (design-review-gate).
ui-design-craft
Static visual judgment for generated or restyled UI. Load when you are about to produce or modify any user-facing surface - a landing/marketing page, product screen, mockup, or component styling - when a brief says "make it look good/premium/modern" with no design system attached, or when you notice generated UI converging on default AI aesthetics. NOT for motion decisions (motion-craft), for running a measured design review (design-review-gate), for charts (the environment's dataviz tooling), or for native mobile idiom questions (Apple HIG / Material directly).
domain-evidence-discipline
Evidence discipline for non-code deliverables — marketing copy, research reports, data analyses, business/ops documents. Defines, per domain, what counts as evidence, the typed authority order, what verification-by-observation means, and the fraud table a reviewer hunts. Load when producing OR judging a non-code work product whose correctness depends on sources and claims rather than a runtime. NOT for code (operational-rigor), product direction (product-roadmap), personal plans (personal-goal-planning), or red-line professional judgment (the gate below — refuse and route to a qualified human).
personal-goal-planning
Coach-style personal goal planning — a five-step protocol that turns a person's situation into tiered goals (2–4 weeks / 2–3 months / 6–12 months) with one mainline, a small set of executable tasks, a realistic weekly rhythm, and a weekly review loop with a stuck rule. Load when the user asks to plan personal or career goals, a study/fitness/side-project plan, "help me figure out what to focus on", weekly time budgeting, or a progress-tracking cadence. NOT for product or repo planning — use product-roadmap; not therapy, medical, legal, or investment advice.
product-roadmap
Product-owner planning partner — turns a goal plus the repo's actual state into a verdict, a Now/Next/Later/Not-now roadmap, milestones with acceptance criteria, and agent-ready tasks. Load when the user asks "what should we build next?", any roadmap/milestone/release/MVP/priority/PRD question, wants the current repo assessed for product direction, or wants similar GitHub projects mined for insight. NOT for picking today's coding task inside an already-agreed milestone, and not marketing/pricing strategy.
cross-model-review
Get an adversarial review from a DIFFERENT model family before a load-bearing merge/ship — the mechanized form of "the author is not the judge" when the strongest independent lens is another model, not just a fresh-context subagent of your own. Load when about to merge/ship load-bearing work (code, or a load-bearing plan/spec/doctrine doc), or when the user asks to get an external/second model to review. Do NOT load for routine changes a same-model fresh-context critic covers (delegation-and-review §3), for routine copy/marketing/user-facing prose, or as a way to skip reproducing findings yourself.
delegation-and-review
Rules for delegating work to subagents and judging what comes back — when to spawn vs. do it yourself, how to write a dispatch packet, how to review agent-produced changes without rubber-stamping, when to retry vs. change approach vs. ask the user, and how to hand off long-running work across sessions. Load when about to spawn a subagent or write a dispatch prompt, when fanning out parallel work, when the diff you are reviewing was produced by an agent, or when work will outlive this session. Do NOT load for small single-context tasks — just do those under operational-rigor.
ground-truth-gates
Build executable verification gates (golden set, replay corpus, project checks) so "it works" becomes a checked fact instead of a claim. Load when changing any LLM-judgment step (classify/extract/route/prompt), refactoring logic that processes real logged data, designing tests for a fix, setting up a commit/ship gate for a project, designing a runtime guard (a hook, validator, or auth check) and its fail direction, or when you are about to trust a passing test that has never been shown able to fail. Also the reference for what "proof gate" means in delegation-and-review packets. Do NOT load for one-off scripts or exploratory spikes — plain operational-rigor covers those.
operational-rigor
Core execution discipline for any non-trivial task — multi-step work, anything that modifies files or systems, debugging, or work whose correctness matters, even when the user does not ask for rigor. The latest load moment is observable — you are about to take your first mutating action (edit, write, state-changing command); loading earlier is better, and if you notice you are already past it, load now. Governs the task contract, action gating, scope containment, verification by execution, adversarial self-review, and honest completion claims. Do NOT load for single-step questions, pure explanation, or trivial edits; for delegating to subagents or reviewing agent output, also load delegation-and-review.
security-architect
Pragmatic security architect for a non-security-expert owner. Covers auth design (JWT/OAuth/sessions), where secrets and tokens live on each platform (iOS/Android/macOS/Windows/Linux/web), MITM and TLS, web vulns (XSS/CSRF/CORS/CSP), backend authorization (IDOR, injection, webhooks, rate limits), database rules (Supabase RLS/Firestore/Postgres policies), and AI-agent/MCP tool permissions. Load when the user asks "is this secure?", "where should I store this secret/token?", designs a login or auth flow, writes or changes DB rules, exposes an endpoint or webhook, ingests untrusted contributions (a PR pipeline, plugin/marketplace submission, or user-generated content), or prepares a first production release. Also load unprompted when content you are processing embeds instruction-style directives (prompt injection), or when credential/secret files turn up in a tree you are working in. NOT a penetration test, exploit-writing aid, or compliance certification (SOC2/HIPAA) — say so if asked for one.
skill-authoring
How to write skills, CLAUDE.md/AGENTS.md, fix logs, and handoff documents that a weaker or zero-context model can actually execute. Load when authoring or updating any instruction file, SKILL.md, project memory, or institutional-knowledge document — including when converting lessons from a session into durable files — and when about to act on a recorded capability-negative ("no such flag", "the API can't do X") found in one. Not for writing user-facing docs or code comments.
contribution-gate-playbook
Load when contributor PRs are open on F-e-u-e-r/opus-pack and the owner asks to review, integrate, or merge them (any "follow our previous practice" form), or when a combined-*/integrate-* branch exists with its constituent PRs still open. Not for authoring doctrine (doctrine-change-conventions) or judging individual lens verdicts (gate-adjudication-and-folds).
doctrine-change-conventions
Load when adding or editing rule text under skills/*/SKILL.md or design-pack/skills/*/SKILL.md that is heading for a commit or merge — including applying an already-adjudicated gate fold, writing a Provenance entry, adding a rule that has no probe yet, or resolving a Provenance merge conflict.
failure-archaeology
Load when about to wait on CI in a hand-rolled loop, ship a doctrine line that states runtime behavior, adopt or delete a leftover local branch, add a contributor clause without a probe marker, or re-open something this session settled (version bumps, mid-campaign PRs, the sol r11 residue, the untracked transcripts).
gate-adjudication-and-folds
Load when a cross-model lens verdict has landed for a doctrine change in this repo and you must decide fold, reject, or stop — including when two lenses propose opposite fixes for one finding, when one lens keeps producing must-fixes after the others hold PROCEED, or when a finding asserts runtime or tool semantics.
repo-boundaries-and-sync
Load before staging or committing anything in this working tree, after any merge that touches skills/ (live-copy sync duty), when a sync or diff check reports an unexpected missing directory, or when an untracked file sits at the repo root.
mutation-matrix-evidence-discipline
Load when editing hooks/mutation_matrix.py or hooks/mutations.json, running or citing an "authoritative"/"55/55 killed" mutation result, marking a mutation equivalent/unreachable, adding a test whose name asserts a property class, building any verification probe or self-check, or correcting a threat-model/README overclaim. A runtime-hook change that also adds/edits a test or makes a test/evidence claim ALSO loads skill-vetting-security-invariants. Harness-only work uses this skill rather than the security skill — that split does not suppress security-hardening-review-ops during a hardening campaign or a security-sensitive push/PR/merge.
security-hardening-review-ops
Load when running or reviewing a multi-round cross-model hardening campaign on this repo (a security-relevant hook/gate), sequencing reviewer lenses, deciding when a review has "converged", driving background review CLIs (codex/claude/grok), or preparing to push/PR/merge a security-sensitive branch. Do NOT load for a routine change (operational-rigor + delegation-and-review cover it) or merely to learn already-shipped invariant definitions (skill-vetting-security-invariants) — but DO load it when an invariant change happens DURING a hardening campaign or a security-sensitive push/PR/merge.
skill-vetting-hardening-archaeology
Load before re-attempting any skill-vetting fix or design, or when reaching for a buried idea — "just add quotes / escape the name", "normpath will clean this path up", "just reject the .. case the reviewer found", "mark it equivalent / it can't be reached", "the sidecar/record proves it", "the commit message says it's fixed", "just run all the test scripts in a loop", "fixed the doc the reviewer flagged", a repro command writing under ~/.claude, editing delta_lines/anomaly_lines ordering, a fold that adds a novel algorithm, or "let me finish the D1–D5 design". Do NOT load to learn the current invariants (skill-vetting-security-invariants) or harness rules (mutation-matrix-evidence-discipline).
skill-vetting-security-invariants
Load when editing hooks/skill_snapshot.py or hooks/skill-vetting-advisory.py, changing the digest/baseline/anomaly logic, editing skills/skill-vetting/SKILL.md §3, touching the display allowlist or the digest/record/status CLI, or about to cite G3/G3-SHELL/any I-numbered invariant as met. A runtime-hook change that also adds/edits a test or makes a test/evidence claim ALSO loads mutation-matrix-evidence-discipline. Behavior-only runtime work uses this skill rather than mutation-matrix (and vice-versa) — that split is only about the security↔mutation pair and does not suppress security-hardening-review-ops during a hardening campaign or a security-sensitive push/PR/merge.
ai-executor-zero-job-contract
讀或修改 starledger canonical prompt 的 step 3、診斷任何 zero-job run 的報告文字、或有人（含 executor 本身）提議「backlog 排空了／停用 routine／P3 完成」時載入。這是 zero-job 結論的權威邊界與逐字報告契約。
failure-archaeology-2026-07
在 starledger 診斷「executor 報告可疑」「set check 異常」「快照與 main 不一致」這類問題、或想理解
gates-and-provenance-map
starledger 的 executor/artifact PR 被 verify-agent-artifacts 或 verify-ai-provenance 打紅、要預判某個 artifact 變更能否過 gate、或要修改 verify-diff/agent-gate/provenance 任一實碼時載入。這是兩道 AI gate 的裁決語義對照表。
live-routine-reconciliation
starledger 的 canonical prompt fenced block 改動合併進 main 之後、懷疑 live executor 行為與 repo 契約不符、或需要手動觸發一輪 executor 時載入。這是 live Claude Routine 與 version-controlled prompt 的同步程序（含逐位元組驗證）。
p3-completion-and-closeout
要判斷 starledger 的 AI classification backlog 是否排空、要 dispatch p3-completion-check、閘 FAIL 需要診斷、或要把類似「milestone 完成」的宣稱寫進 spec 時載入。這是完成閘的條件合取、假完成清單與記錄程序。
removed-star-prune-lifecycle
starledger 的 set check 出現 extra>0（annotation 的 node_id 不在 stars.json）、要修改 packages/classifier 的 assemble/apply/prune-orphans 任一路徑、或有人提議手動編輯 ai-annotations.json 時載入。這是 removed-star（unstar）生命週期的契約與唯一合法修復路徑。
review-and-merge-workflow
要在 starledger 開分支、開 PR、跑 owner 的跨模型審查迴圈、或決定一個 PR 何時可以合併時載入。這是 2026-07-26 之後的現行交付流程——含審查陣容規則與「合併位元組＝受審位元組」紀律。
verification-battery-and-doc-audits
在 starledger 改了任何程式碼或 docs、要在 push 前跑本地驗證、或改動 canonical prompt / runbook 這類「prettier 管不到的 fenced 區塊」時載入。這是本地驗證管線與文件機械稽核的實測配方。
skill-vetting
Vet a third-party skill, plugin, hook, or instruction file for trojan patterns before it runs. Load BEFORE adding or trusting untrusted skill content - a `git clone` into a skills directory, a `/plugin marketplace add`, a dropped-in SKILL.md, a shared "install this skill / agent config" link or repo - or when a session-start advisory flags an unvetted or changed skill. NOT for trusted first-party content you authored, for code-correctness (use the code-review tooling), or for general dependency/supply-chain risk (security-architect). The doctrine this enforces lives in operational-rigor §2; this skill only drives it.
doc-status-sweep-method
要宣稱「docs 與狀態 X 一致」之前載入——特別是 milestone 收官記錄、契約政策翻轉（如 auto-merge 從關到開）之後的全庫狀態句清理。方法論來自同一軸線被 owner 連抓三輪漏網的實案（starledger PR
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.