nwiizo
10 verified skills890 total stars
.claude/skills/benchmark-runner
# Benchmark Runner Workflow Multi-step workflow for running performance benchmarks on ccswarm. ## Overview This skill guides you through running, analyzing, and comparing performance benchmarks for the ccswarm system. ## Benchmark Types ### 1. Microbenchmarks Fine-grained performance measurements for specific functions. ### 2. Integration Benchmarks End-to-end performance for complete workflows. ### 3. Load Testing System behavior under sustained load. ## Running Benchmarks ### 1. Setup
testing135
.claude/skills/hitl-approval
# HITL (Human-in-the-Loop) Approval Workflow Multi-step workflow for human approval integration in ccswarm agent operations. ## Overview This skill guides you through implementing and using human-in-the-loop approval mechanisms for high-risk agent operations. ## When HITL is Required ### High-Risk Operations - File deletions - Database modifications - External API calls - Configuration changes - Deployment actions ### Risk Levels | Level | Action | HITL Required | |-------|--------|-------
development135
.claude/skills/git-worktree
# Git Worktree Workflow Multi-step workflow for parallel development using git worktrees. ## Overview Git worktree allows working on multiple branches simultaneously. Each worktree is an independent working directory with its own branch. ## Setup Workflow ### 1. Create Feature Worktree ```bash git worktree add ../ccswarm-feature-<name> feature/<description> ``` ### 2. Create Bug Fix Worktree ```bash git worktree add ../ccswarm-bugfix-<name> hotfix/<description> ``` ### 3. Create Experimen
development135
.claude/skills/rust-agent-specialist
# Rust Agent Specialist Workflow Apply Rust-native patterns to ccswarm codebase development. ## Overview This skill provides guidance for implementing Rust-idiomatic patterns in the ccswarm multi-agent orchestration system. ## Core Patterns ### Type-State Pattern Compile-time state validation with zero runtime cost. ```rust // State types pub struct Uninitialized; pub struct Initialized; pub struct Running; pub struct Agent<State> { inner: AgentInner, _state: PhantomData<State>,
development135
.claude/skills/deploy-workflow
# Deploy Workflow Multi-step workflow for deploying ccswarm releases. ## Overview This skill guides you through the complete deployment process for ccswarm releases, from building to publishing. ## Pre-Deployment Checklist ### 1. Version Update ```bash # Update version in Cargo.toml files # Root workspace # crates/ccswarm/Cargo.toml # crates/ai-session/Cargo.toml ``` ### 2. Quality Gates ```bash # Run full quality check cargo fmt --all cargo clippy --workspace -- -D warnings cargo test --w
tools135
echo-skill
Use when the user provides an arbitrary line of text and you must echo it back verbatim, prefixed with "ECHO:".
tools43
zap-triage
Turn OWASP ZAP JSON reports into code-level remediation work for any authorized web application without launching unscoped scans.
development43
tools/vigil/skills/owasp-assessment
# OWASP Assessment — 詳細仕様 2つの OWASP 標準に基づく網羅的セキュリティ検査。 - **OWASP Top 10:2021** — Web アプリケーション向け（A01〜A10） - **OWASP API Security Top 10:2023** — API 向け（API1〜API10） 各カテゴリに対して: 検査項目、CWE マッピング、grep パターン、判定基準、Opus 4.6 による深掘りポイントを定義する。 --- # Part 1: OWASP Top 10:2021（Web アプリケーション） 公式: https://owasp.org/Top10/ ## A01:2021 — Broken Access Control **概要:** アクセス制御の不備。ユーザーが許可された範囲を超えて操作できる。2021年版で1位に上昇。テスト対象の94%で検出。 **主要 CWE:** - CWE-200: 機密情報の未認可アクターへの露出 - CWE-201: 送信データへの機密情報の挿入 - CWE-352: CSRF -
tools43
contests/juice_shop/.claude/skills/playwright-attack
# Playwright Attack Patterns Juice Shop を Playwright MCP で攻撃するパターン集。 ## SQLi ログイン ``` 1. browser_navigate → http://localhost:3000/#/login 2. browser_snapshot → ref確認 3. browser_type → email: "' OR 1=1--" 4. browser_type → password: "a" 5. browser_click → Loginボタン ``` ## XSS 攻撃 ``` browser_navigate → http://localhost:3000/#/search?q=<iframe src="javascript:alert('xss')"> ``` ## API 操作 (fetch) ```javascript browser_evaluate → function: () => fetch('/api/Users', { method: 'POST', headers
tools43
memory-optimizer
Refactors CLAUDE.md into minimal startup context by extracting path-specific rules, skills, commands, and agents. Use when CLAUDE.md exceeds 50 lines, startup feels slow, memory needs restructuring, or splitting monolithic project instructions.
development43