Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
Skill Claude Code
Full context window audit for Claude Code workspaces. Analyzes CLAUDE.md files, MEMORY.md + individual memory files, hooks, and root directory for token bloat, stale entries, orphans, redundancies, and over-firing hooks. Generates actionable report with estimated token savings. Use when the user says "context audit"…
Skill Claude Code
Work through a PR's unresolved review comments one at a time. Detects the repo's stack, fetches the unresolved threads, reproduces each reviewer's claim before touching anything, fixes only what the probe confirmed, proves the fix with the same probe, presents it, then commits that one thread before moving on.
Skill Claude CodeCodex
VEDA first-time setup. Conversational interview that fills in CLAUDE.md and MEMORY.md placeholders so the user does not have to edit those files manually. Use when the user opens VEDA for the first time, when placeholders like [YOUR NAME] are still present in CLAUDE.md, or when the user says /onboard, "onboard me"…
NiroDu/claude-code-harness-kit
Skill Claude CodeCodex
Use when a repository works with claude-code-harness-kit and you want Claude to restore state, initialize or validate the harness, ignore stale session context, and continue work through the repo artifacts instead of guessing from memory.
Skill Claude Code
Writes, audits, and continuously evolves project memory across the whole Claude Code memory system — CLAUDE.md (project ./CLAUDE.md, personal /.claude/CLAUDE.md, local ./CLAUDE.local.md), AGENTS.md, .claude/rules, and Auto Memory. Use when the user wants to create, generate, review, audit, slim down, or evolve a…
Skill Claude CodeCodex
Backend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js API routes.
justinjia0813/digital-person-skill
Skill Claude Code
A tool that studies someone’s public posts on WeChat public accounts, Weibo, and Twitter/X to extract their ideas, decision methods, and writing style.
rodrigo-kiko/claude-reflect-skill
Skill Claude CodeCodex
Analyzes development sessions to extract learnings from corrections, successful patterns, and revealed preferences. Use when the user wants to reflect on a coding session, extract lessons learned, or update project knowledge. Triggered by /reflect command or phrases like "reflect on this session", "what did we learn…
Skill Claude Code
Top-down, whole-system audit of a codebase's DEEP LOGIC - the bugs that live in the SEAMS between modules, not inside any one file. Finds contracts that drift across module boundaries, invariants enforced on one path but broken on a sibling path, cross-cutting responsibilities (deletion cascade, tenant-scope…
michaelcjoseph/agent-coding-setup
Skill Claude Code
Run the full review panel — test-specialist, security-auditor, code-reviewer, code-simplifier, and architecture-reviewer — in parallel against the current uncommitted working tree, then return one consolidated verdict.
Skill Claude Code
Deep-mines the last 50 sessions of a project to find recurring corrections, repeated instructions, and reverted changes. Writes proposed CLAUDE.md rules to CLAUDE.md.undrift — review and merge manually. Run when patterns keep repeating or after /undrift-check flags a project as needing attention.
Skill Claude Code
Split instruction files (CLAUDE.md / AGENTS.md) into skills and keep them drift-free. Use when asked to audit docs, split docs, lint docs, reorganize documentation, or extract skills.
Skill Claude Code
This skill should be used when the user asks to "review my mod", "check my code", "code review", "audit the plugin", "/mc-review", or requests a quality check on Minecraft mod/plugin code. It performs static analysis for correctness, performance, compatibility, and best-practice issues specific to Minecraft Forge…
Skill Claude CodeCodex
Produce publication-quality scientific figures in a restrained, vision-science-inflected house style. Use this skill whenever the user asks for plots, figures, or visualizations for a paper, preprint, talk, or grant — especially for vision, psychophysics, neuroscience, or computational modeling work — and any time the…
Skill Claude Code
Switch to freelance closing mode — mission qualification, scoping, quoting, and client relationship management.
VictorCostaOliveira/agents-tdd-flow
Skill Claude Code
Creates specifications before code. Phase 1 (Specify) is always delegated to the requison subagent — it elicits requirements and drafts content aligned to this skill’s template. Use when starting a project, feature, or large change without a spec. With Figma, run the MCP flow (below) before locking the spec. Suggested…
lucas-antunes/growth-ops-blueprint
Skill Claude CodeCodex
Use quando o usuário pedir uma análise completa de performance, check semanal, check diário, ou "como está {{COMPANYNAME}}". Cobre todas as seções do negócio em modo diário ou semanal.
Skill Claude CodeCodex
Evaluate prompts, skills, and Claude Code agents for correctness and optimization against the latest Claude best practices. Use when the user asks to "evaluate," "review," "audit," "optimize," "modernize," or "upgrade" a prompt, system prompt, SKILL.md, agent definition, or CLAUDE.md, or says things like "I wrote this…
Skill Claude CodeCodex
Use this when the user wants a summary of a milestone. This is a speed optimized skill that does not do any git actions, so it may be stale. Use /milestone-summary instead for a fresh report.
Skill Claude CodeCodex
Generate a project-specific AGENTS.md (with CLAUDE.md as exactly @AGENTS.md) with coding standards, style guides, and best practices. Triggers on any request to set up a project, initialize coding standards, create or update CLAUDE.md or AGENTS.md, configure agent instructions, add code quality rules, or start a new…
Skill Claude CodeCodex
End-of-session self-improvement. Review the session for user corrections, steering, and reusable discoveries, then route them into global or project CLAUDE.md, auto-memory, GSD planning docs, or (rarely) a skill, with a compact preview and per-item approval before anything is written. Use when the user invokes…
gmen1057/claude-code-starter-kit
Skill Claude Code
Use when about to claim work is complete or fixed — requires running verification commands and confirming output before making any success claims.
Skill Claude CodeCodex
Cria a stack completa de engenharia confiável (spec + CLAUDE.md + hooks + regras modulares) em qualquer projeto. Use quando o usuário pedir para inicializar o harness, montar a estrutura da aula, subir o controle antes de codar, ou preparar um projeto para desenvolvimento assistido por agente. Materializa .spec.md…
Skill Claude Code
Deep research and knowledge synthesis. Activates for ANY task involving factual claims, technical decisions, implementation patterns, or domain knowledge. Deactivated ONLY when the mechanical task test passes (all three conditions true).
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: