All authors
Claude Skills by qyndex
github.com/qyndex62 skills0 installs8 views
- CommandsCreate a new skill from a description (or from instinct/signature evidence). Generates to .claude/memory.proposed/skills/ for review via /dream-review --approve-skill. Round 9 A.Votes: 0GitHub stars: 2
- AnalyzeCross-check spec, plan, and tasks for consistency. Phase 4.5 of the eight-phase workflow. Catches scope drift, missing acceptance criteria, dependency cycles. Modeled on github/spec-kit /speckit.analyze. Run after /tasks, before /implement.Votes: 0GitHub stars: 2
- Animation DisciplineUse when adding transitions, hover states, scroll triggers, motion. Enforces easing/duration discipline, prefers-reduced-motion, GPU-friendly properties only, View Transitions API for routes. Round 9 C.Votes: 0GitHub stars: 2
- Api VersioningManage public API version transitions — v2-alongside-v1, Deprecation/Sunset headers, client migration tracker. Use when a breaking API change is needed but cannot ship as a wire-break.Votes: 0GitHub stars: 2
- AutopilotFive-phase autonomous execution. Spec expansion → planning → execution → QA loops (max 5) → multi-perspective validation → cleanup. Aborts if same error repeats 3 times. Used by Cloud Routine overnight runs and any /loop autonomy.Votes: 0GitHub stars: 2
- Browser E2eBrowser-driven E2E discipline. Thin pointer — defers to anthropics:webapp-testing plugin for Python Playwright scripts and to playwright MCP / chrome-devtools MCP / Claude in Chrome based on the situation. Use when verifying any web change.Votes: 0GitHub stars: 2
- CharacterizeWrite characterization (approval/golden-master) tests that pin the CURRENT behavior of untested legacy code BEFORE it is changed. The "no tests = no writes" safety net for brownfield adoption — distinct from tdd-loop (which drives NEW behavior red→green). Use during /adopt Phase 4, or whenever a task must touch a file flagged in .claude/state/adopt/uncharacterized-paths.txt.Votes: 0GitHub stars: 2
- ClarifyWalk through every open question in a spec and resolve it with the user. Phase 2.5 of the eight-phase workflow. Modeled on github/spec-kit's /speckit.clarify. Run after /specify if any [OQ] tags remain.Votes: 0GitHub stars: 2
- CodemodLarge-scale refactor via AST-aware transforms — jscodeshift (JS/TS), ts-morph (TS), comby (any language), bowler (Python). Dry-run → review → apply loop. Use for React 18→19, Node 22→24, or any change touching 50+ call sites.Votes: 0GitHub stars: 2
- Constitution CompactQuarterly constitution compaction. Identifies redundant, stale, or over-verbose sections in .claude/CLAUDE.md, proposes a trimmed version under .claude/memory.proposed/constitution-diff.md, and enforces the ≤300-line cap. Never edits the constitution in-place — human applies the diff.Votes: 0GitHub stars: 2
- ConstitutionAuthor or amend the project constitution at .claude/CLAUDE.md. Use when starting a brand-new project, when a major architectural principle shifts, or when the team wants to lock in non-negotiables. Modeled on github/spec-kit's /speckit.constitution.Votes: 0GitHub stars: 2
- Context BudgetInventory and recommend prunes for loaded agents, skills, MCP servers, and rules. Reports always-needed / sometimes-needed / rarely-needed buckets with token costs. From affaan-m/ECC context-budget.Votes: 0GitHub stars: 2
- Creative WebUse when a spec EXPLICITLY warrants "wow factor" beyond default Tailwind — parallax, 3D, generative art, scroll-triggered storytelling. Hard-guardrailed by Lighthouse perf ≥ 90, a11y = 100. Round 9 C.Votes: 0GitHub stars: 2
- Db MigrationExpand/contract pattern for zero-downtime database schema changes. Replaces the (wrong) \"forward-only by default\" rule. The four-phase pattern: expand → backfill → cutover → contract.Votes: 0GitHub stars: 2
- DebugRun the four-phase systematic debugging loop (reproduce → isolate → diagnose → fix). Delegates to debugger agent. Modeled on superpowers/systematic-debugging.Votes: 0GitHub stars: 2
- Detect StackDetect the project's tech stack(s) — frameworks, ORMs, build tools, test runners — from lockfiles and config files. Returns a structured manifest. Used by atlas-refresh, lint dispatcher, stale-deps workflow. Round 8 C.Votes: 0GitHub stars: 2
- Dispatch CriteriaSix gates for deciding whether to delegate to a subagent or do it in-session. From ChristopherKahler/paul. The "is this work parallelizable" decision aid.Votes: 0GitHub stars: 2
- DreamMemory consolidation. Dedupe and compress .claude/memory/ topic files, absolutize relative dates, archive stale instincts and processed checkpoints, and hard-enforce the 200-line MEMORY.md cap. Produces a reviewable proposal, never a silent overwrite.Votes: 0GitHub stars: 2
- ExperimentA/B test design — hypothesis, MDE, sample size, primary/guardrail metrics, significance gate. Different from flag-rollout (which gradually exposes to ALL users). An experiment runs two variants in parallel to learn which is better.Votes: 0GitHub stars: 2
- FastapiUse when project is FastAPI (Python async web framework). Covers Pydantic v2 models, dependency injection via Depends, async SQLAlchemy 2.x, lifespan context manager, BackgroundTasks vs Celery, routers per feature. Round 9 B.Votes: 0GitHub stars: 2
- Flag CleanupRemove a feature flag from code after it's been 100% for the cleanup_after window. Greps call sites, replaces `if (flags.X)` branches with the new path, deletes the flag from the provider, archives the spec/initiative reference.Votes: 0GitHub stars: 2
- Flag RolloutPhased rollout discipline for production. Ship code default-OFF, ramp 1%→10%→50%→100% with monitoring windows, auto-rollback on metric breach, scheduled cleanup. The "phased delivery in prod" backbone.Votes: 0GitHub stars: 2
- Flask RealtimeConventions for a Flask REST API with WebSockets (Flask-SocketIO). Invoke when atlas shows `flask` in pyproject/requirements and any of `flask-socketio`, `eventlet`, `gevent`, `python-socketio`. Covers blueprint structure, SocketIO room patterns, JWT+Flask-Login auth, SQLAlchemy+Alembic migration discipline, Celery/RQ background jobs, pytest fixtures, and the eventlet/gevent/ASGI decision.Votes: 0GitHub stars: 2
- Framework DetectorMaps a detected stack (from `detect-stack` / atlas) to the framework-specific SKILL.md that should be activated for this session. Invoke once at session start, or whenever a new spec mentions a stack that isn't already loaded. Reads `.claude/memory/atlas/STACK.md` and emits a list of skills to consult.Votes: 0GitHub stars: 2
- Front End DesignUse when designing a new UI, redesigning a screen, building a landing/marketing page, or making something look better. Orchestrates the visual-craft skills (tailwind-discipline, typography, animation-discipline, image-generation, creative-web). Round 9 C.Votes: 0GitHub stars: 2
- Grill MeUse when authoring an initiative, clarifying interdependent open questions, or designing a multi-quarter roadmap — depth-first one-at-a-time interrogation with agent recommendation + reasoning per question. Skip for bug-fix / implement / verify (too much friction).Votes: 0GitHub stars: 2
- HandoffStructured handoff between subagents, between background sessions, or between the night-run and the morning session. Uses the NEXUS YAML schema (v1.0). Required for swarm streams; recommended for all other subagents. Includes attempt counter, evidence requirement, explicit next-step.Votes: 0GitHub stars: 2
- Image GenerationUse when a user-facing visual asset is needed (hero, illustration, icon, OG card, badge, mock). Orchestrates reuse-first → programmatic → icon → photo → AI illustration → vector → text-in-image. Round 9 C+D.Votes: 0GitHub stars: 2
- ImplementExecute tasks from tasks/TASKS.md one at a time using strict TDD. Phase 5 of the eight-phase workflow. Delegate each task to the implementer agent. Walks the DAG, respects deps, fans out [P] tasks if requested.Votes: 0GitHub stars: 2
- InitiativeAuthor or amend a multi-quarter initiative — the program-level artifact above per-feature specs. Use for 3+ month efforts spanning multiple specs, multiple teams, or multiple quarters. An 18-month migration is one initiative; a 2-day feature is not. Spec template's `complexity: XL` should split into an initiative + multiple specs.Votes: 0GitHub stars: 2
- InstinctContinuous-learning observer. PostToolUse + Stop hooks write JSONL observations; a background Haiku subagent extracts atomic "instincts" (trigger → action, confidence-rated, domain-tagged). Project-level by default; promotes to global at 2+ projects. From affaan-m/ECC continuous-learning-v2.Votes: 0GitHub stars: 2
- LoopRun a continuous Ralph-style autonomous loop until a budget or success condition is hit. Each iteration restarts with a fresh subagent that reads PLAN.md + progress.md from disk, picks the next unblocked task, executes, verifies, and writes back. Native to Claude Code via /loop, but this skill formalizes the discipline.Votes: 0GitHub stars: 2
- NextjsUse when project is Next.js (App Router or Pages Router). Covers Server vs Client Components, Server Actions, Suspense boundaries, route handlers, middleware, next/image discipline, revalidation. Round 9 B.Votes: 0GitHub stars: 2
- ObservabilityScaffold structured logs, metrics, and trace spans for any new endpoint or service. Verify emissions in PR diffs. Enforce the observability rules from .claude/rules/backend.md mechanically (not just aspirationally).Votes: 0GitHub stars: 2
- Okr AlignCheck that a spec or initiative aligns with a current-quarter OKR. Run automatically during /specify (Phase 2) and /initiative create. Flags work that has no KR backing.Votes: 0GitHub stars: 2
- Parallel SwarmCoordinator-led parallel feature development. 5-10 features built simultaneously via native `claude --bg --worktree feat-<N>`. Each stream has own subagent team; coordinator merges. Built on native CC v2.1.144+ primitives — no custom orchestrator.Votes: 0GitHub stars: 2
- PivotCapture a mid-project pivot (drop, pause, supersede, stop a direction) into pivots/active/<slug>.md with a resolvable target (spec or initiative id) and target_kind. Records trigger, Lean-Startup pivot type, old/new direction, and cascade impact so the change is traceable, not silent.Votes: 0GitHub stars: 2
- PlanAuthor the technical plan for a spec. Phase 3 of the eight-phase workflow. Produces plans/active/<id>-<slug>.md with architecture, data model, API contracts, dependencies, phasing, risks. Delegate to the architect agent. Modeled on github/spec-kit /speckit.plan.Votes: 0GitHub stars: 2
- Pr BodyWrite or refresh a pull request title and body for this harness, including approved spec references, complete acceptance evidence, actual validation results, and remaining risks. Use before opening a PR or updating its scope or proof.Votes: 0GitHub stars: 2
- Prime DisciplineReference card of discipline patterns from across the research — Boris Cherny's CLAUDE.md / compaction / rewind rules, Karpathy's surgical-changes rule, Lovable's good/bad example format, Cursor's semantic-search-first, Devin's plan↔execute toggle. Single source of habits. Invoke manually via /prime-discipline; the skill-router will also hint it on `prime me`, `load context`, `refresh state`.Votes: 0GitHub stars: 2
- Prod ObservabilityScaffold the PRODUCTION observability + alerting + notification stack into a shipped app — OTEL collector, Sloth SLO burn-rate rules, synthetic monitor, Alertmanager routing, Sentry release tracking. Distinct from the `observability` skill (which emits telemetry in app code) — this gives that telemetry a destination + makes SLOs fire alerts. Round 12 A.Votes: 0GitHub stars: 2
- React NodeConventions for a React frontend talking to a Node.js backend (Express / Fastify / Hono). Invoke when the atlas shows both a React web framework AND a Node API in the same repo or sibling repos, or when designing the frontend↔backend contract, auth flow, file upload, or WebSocket upgrade. Covers REST-vs-tRPC, JWT-in-httpOnly-cookie rotation, multipart upload, ws handshake, SSR-vs-CSR decision.Votes: 0GitHub stars: 2
- React ViteConventions, pitfalls, and canonical patterns for a React + Vite SPA. Invoke when editing files under a project whose atlas detected `vite.config.{js,ts,mjs}` + React, or when the user explicitly asks for React+Vite advice. Covers component patterns, state-management decisions, Vitest+RTL+Playwright testing, build optimization, and the top hooks pitfalls.Votes: 0GitHub stars: 2
- Release TrainCadenced release process — weekly cut on Friday, staging Monday, canary Wednesday, 100% Friday next. Different from release-on-demand. Use when shipping multiple PRs together with predictable timing.Votes: 0GitHub stars: 2
- ResearchGround a decision in evidence. Delegates to researcher agent. Produces a citation-heavy comparison brief saved under docs/research/. Use when a non-trivial choice (library, pattern, API) needs evidence beyond what the codebase shows.Votes: 0GitHub stars: 2
- ReviewRun code review and (optionally) security review on the current diff. Phase 7 of the eight-phase workflow. Delegates to reviewer + security agents. Multi-pass — each finding is re-verified to filter false positives (Anthropic pattern).Votes: 0GitHub stars: 2
- Security GuardProject security discipline. Thin pointer that wires up the semgrep MCP, CodeQL workflow, dependency audits, and the OWASP+LLM Top 10 checklist. Used by the security agent and triggered automatically when diffs touch auth/data/network/deps.Votes: 0GitHub stars: 2
- Self HealWhen a test fails, lint errors, or CI breaks — diagnose with debugger agent, propose minimal fix, implementer applies, verifier reruns, loop up to 3 attempts then escalate. Combines Anthropic Auto-Fix + Playwright self-heal + 4-tier escalation pattern.Votes: 0GitHub stars: 2
- ShipFinal phase — push branch, open PR, wait for CI, merge, deploy. Phase 8 of the eight-phase workflow. Delegates to release agent. Refuses to proceed if any quality gate fails.Votes: 0GitHub stars: 2
- Skill CreatorUse when a recurring task pattern emerges that deserves its own skill. Generates a fresh SKILL.md under .claude/memory.proposed/skills/<slug>/ (NOT activated until /dream-review --approve-skill). Triggers — manual (/create-skill), instinct promotion (--as-skill on a mature instinct), or task-signature-detector (3+ Bash sequence repeated ≥5× across ≥2 sessions). Round 9 A.Votes: 0GitHub stars: 2