Harness Govern — Security skill for Claude Code
End-to-end harness governance loop and single front door for the harness- family — measure (harness-budget + harness-audit) -> conform to official Claude Code best-practices -> fix (harness-revise).
How to install Harness Govern
Installs to ~/.claude/skills/hmj1026-dhpk-harness-govern/SKILL.md
mkdir -p ~/.claude/skills/hmj1026-dhpk-harness-govern && curl -fsSL https://raw.githubusercontent.com/hmj1026/dhpk/HEAD/commands/harness-govern.md -o ~/.claude/skills/hmj1026-dhpk-harness-govern/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Harness Govern does
description: 'End-to-end harness governance loop and single front door for the harness-* family — measure (harness-budget + harness-audit) -> conform to official Claude Code best-practices -> fix (harness-revise) -> verify. Read-only by default; --fix opts into mutations. Safe to /loop.' argument-hint: '[--fix] [--scope repo|skills|rules|mcp]' allowed-tools: 'Read, Grep, Glob, Bash, Skill' disable-model-invocation: true metadata: dhpk-invocation-class: explicit-only
/dhpk:harness-go
Alternatives in Security
- Node9 Proxy — The Execution Security Layer for the Agentic Era 209 ★
- Govern Audit — /govern-audit 145 ★
- Sovereign Skills — 20 production-grade skills for AI coding agents — setup, scope, discipline, code review, security, session man 125 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Front Door Security
Front-door security pass — find the externally-reachable and irreversible exposures, fix what's safe to fix, f
MCP Govern
Run MCP governance audit — check 30 rules across security, access control, compliance, data governance, and op
Raindrop Collection Governance
Agent Skills-compatible skill: govern a Raindrop.io library's collection structure via raindrop-mcp — audit, c
SEO God
Turn Claude Code into an agentic SEO operator for your own site — audit, measure, daily loop, AI visibility, s
Vibe Guard
Full safety check for AI-generated code before pushing. Runs production resilience, security, and comprehensio
Agentic Security Pipeline
An autonomous agentic pipeline that finds, proves, and patches real C memory-safety vulnerabilities end-to-end
Related Agents
Harness Orchestrator
Orchestrator — owns ONE feature end to end at layer 1. Runs the loop: delegate to leads, assess team digests,
Jobseeker
The single front door for ALL job-search tasks — address it as "@jobseeker" or "jobseeker". Use for anything j
Aim
Internal AIM orchestration helper used by the AIM project skill; not the user-facing front door.