meta-context-diet
Audit CLAUDE.md and .claude/rules for always-loaded content that should become an on-demand skill. Use when CLAUDE.md feels bloated, trimming context, or promoting a rule into a skill.
plugin-authoring
Add, modify, or delete a skill or plugin in this repo — frontmatter shape, the seven metadata files a plugin touches, and the create/delete checklists. Use when creating a new skill or plugin, removing one, or asking which files a plugin change must update.
cc-only-skill
Configure Claude Code hooks in settings.json. Use when authoring PreToolUse hooks.
normal-skill
Commit staged changes with a conventional message. Use when saving work to git history.
folded-description
>-
quoted-description
Debug HTTP APIs: trace requests, inspect headers. Use when a request fails: check status first.
adversarial-review
Adversarial second-pass review that tries to break code, designs, plans, or ADRs. Use when stakes are high and a normal review already ran.
agent-cli-worktree-safety
Data-loss invariants for a CLI that runs an agent SDK in a git worktree then force-removes it. Use when building or reviewing such a CLI, or when an agent run reports "no changes" after writing files.
cold-read-gate
Gate outward-bound text (upstream issues, docs, PR bodies) through isolated haiku fresh-reader critique before publishing. Use when an artifact must survive a reader with zero project context.
exclusive-lock-dispatch
Pre-dump-then-dispatch for tools holding an exclusive lock (Ghidra, migrations, single-writer caches). Use when fanning out parallel agents needing a non-concurrent resource.
execution-grounded-review
Execution-grounded review: run tests first, trace each acceptance criterion to execution evidence. Use when verifying an implementation meets spec.
expose-tunable-knob
Expose a live-adjustable control instead of guessing a magic constant for a parameter the agent cannot itself perceive. Use when tuning visual, audio, or UX output (color, size, position, timing, volume) that only the user can judge.
harness-tool-errors
Claude Code harness tool errors — Read on a missing path, directory or oversized file; Edit before Read or after a formatter; Bash permission denials. Use when a Read, Edit, Write or Bash call errors.
mcp-server-authoring
FastMCP authoring for Python MCP servers — tools, resources, prompts, TDD, release-please. Use when building or extending an MCP server, adding a tool/resource, or scaffolding a new server.
tool-result-traps
Tool results that mean something other than they look — a pattern that never compiled, a glob that cannot match a directory, one tier searched but reported absent, a rejected flag reading as "no results". Use when an empty result is about to gate an action, land in a doc, or be reported as done.
verify-before-plan
Verify orchestrator premises (file counts, build state, artefact presence) before dispatching parallel subagents. Use when a wave's briefs cite a number, path, or behaviour not yet checked.
wave-based-dispatch
Sequential-wave dispatch for WO chains where output of one feeds the next, shared locks, or shared files prevent fan-out. Use when planning dependent multi-WO landings.
bevy-ecs-patterns
Advanced Bevy ECS: complex queries, system scheduling, change detection, and performance tuning. Use when optimizing Bevy architecture or implementing complex game systems.
bevy-game-engine
Bevy game engine: ECS, rendering, input, and asset management. Use when building Bevy games, working with entities/components/systems, or mentioning Rust gamedev or 2D/3D games.
blog-post-writing
Style guide for technical blog posts — updates, retrospectives, tutorials, deep dives. Use when writing about work done, documenting a project, or mentioning blog/post/devlog.
adr-relationships
Domain analysis, conflict detection, and relationship validation for Architecture Decision Records. Use when creating or validating ADRs to ensure consistency.
blueprint-autonomy-level3
Install blueprint autonomy level 3 (ADR-0020): scheduled-autorun + approved-work-order-execution GitHub workflows. Use when enabling out-of-band blueprint automation in a repo.
blueprint-autopilot
Run due blueprint maintenance ambiently at autonomy level 2+. Use when a drift nudge reports blueprint tasks due or suggests /blueprint:autopilot.
blueprint-migration
Versioned migration procedures for blueprint format upgrades (v1.x to v3.3). Use when blueprint-upgrade needs version-specific logic, content hashing, or rollback.
blueprint-story-audit
Audit user stories against codebase and tests for tier-ranked coverage gaps. Use when running story audit, PRD reconciliation, or surfacing PRD-code drift.
blueprint-story-reconcile
Reconcile PRD requirements with a story-audit drift report. Use when marking PRD entries implemented/partial/missing, or promoting code-only stories into the PRD.
blueprint-workspace-scan
Discover child blueprint workspaces and refresh the manifest. Use when adding/removing a child blueprint or when status shows stale portfolio data.
bulk-sweep-classify
Classify every regex match before a bulk find-replace/syntax-modernization sweep — four match categories, scoped transform, allowlist-aware verification. Use when bulk-renaming commands/tools/paths or migrating syntax across many files.
code-complexity
Analyze code complexity metrics (cyclomatic, cognitive, function length, coupling). Use when identifying refactoring targets, tracking codebase health, or reviewing large changes.
code-dead-code
Detect dead code, unused exports, unreachable branches, and orphaned files. Use when reducing maintenance burden, cleaning up after refactors, or auditing codebase health.
code-dep-audit
Audit dependencies for security vulnerabilities, outdated packages, and license compliance. Use when checking supply chain security, preparing releases, or responding to CVEs.
code-docs-quality
Analyze docs quality across PRDs, ADRs, PRPs, CLAUDE.md, .claude/rules/. Use when auditing documentation, checking for stale ADRs/PRDs, or validating rule frontmatter structure.
code-hidden-failures
Scan for hidden failures: swallowed errors (empty catch, || true, 2>/dev/null) and silent degradation (success on zero results). Use when failures vanish or success masks empty output.
code-license-position
Check a vendor's licence FAQ and discussion threads before calling a LICENSE restriction a blocker. Use when a model or dependency licence seems to forbid use.
code-lint
Run the project's linters and formatters — auto-detects ruff/eslint/clippy/gofmt. Use when linting code, auto-fixing lint errors, or checking formatting.
code-scaffold-backport
Back-port an instance fix into the generator that produced it — cookiecutter, scaffold, `new-*` recipe. Use when fixing, reviewing, or auditing a generated file.
code-test-quality
Analyze test quality: smells, empty assertions, flaky patterns, coverage gaps. Use when tests are unreliable, coverage is misleading, or after major refactors.
comfy-cli
>-
comfy-conditionals
>-
comfy-corpus-validation
>-
comfy-debug-preview
>-
comfy-flow-control
>-
comfy-image-utils
>-
comfy-math-strings
>-
comfy-metadata
>-
comfy-node
>-
comfy-registry-lifecycle
>-
comfy-subgraphs-app-mode
>-
comfy-workflow-json
>-
comfy-workflow-layout
>-
Page 435 of 1657 · 82816 results