Optimize Harness — Security skill for Claude Code
Audit one or all supported agent harnesses for config efficiency, hook overhead, permission friction, plugin and MCP cost, and measured session friction.
How to install Optimize Harness
Installs to ~/.claude/commands/phatblat-dotfiles-optimize-harness.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/phatblat/dotfiles/HEAD/.claude/commands/optimize-harness.md -o ~/.claude/commands/phatblat-dotfiles-optimize-harness.md Restart Claude Code, or start a new session, for it to be picked up.
What Optimize Harness does
name: optimize-harness description: Audit one or all supported agent harnesses for config efficiency, hook overhead, permission friction, plugin and MCP cost, and measured session friction. argument_hint: "[harness: claude|codex|opencode|pi|omp|antigravity|cursor|grok ...] [focus: config|hooks|permissions|plugins|mcp|context|sessions|friction|all] [since:DAYS]"
Load and follow the shared skill at `~/.agents/skills/optimize-harness/SKILL.md`.
Arguments: $ARGUMENTS
Alternatives in Security
- Security Guidance — Hook that warns about command injection, XSS, and unsafe patterns 14k ★
- Security Guardian — Comprehensive security analysis for RTK CLI tool, focusing on command injection, shell escaping, hook security 11.9k ★
- Reflexion — Audit and optimize the .claude/ ecosystem — skills, hooks, guides, patterns, agents, and settings — for qualit 224 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
AI Spend
Audit Claude Code token efficiency: MCP overhead, session patterns, skill cost, cache hit rate. Add --full for
Audit Harnesses
Audit AI Harness Consistency
Auth Security UX Specialist
Use when a login or security flow feels either unsafe or full of friction. Signup, password reset, 2FA and pas
Mine Permissions Audit
Analyze frequent permission prompts and recommend allow-list entries to reduce friction.
Optimizing Skills Performance
Audit and optimize AI agent skills for token efficiency using progressive disclosure principles
Audit Policies
Analyze a project's Claude Code permission settings and suggest toolgate policies. Use when the user asks to a
Related Agents
Harness Radar
Weekly movement briefing on the AI agent harness space — what climbed, what entered the radar, what died — fro
Harness Architect
Expert in designing and building agent harnesses — agent/skill/tool decomposition, subagent teams, hooks, MCP
Cr Data Layer
Data-layer lens — measured query cost, schema migrations and bulk data writes. Spawn when the diff adds or cha