jhlee0409/omni-harness-kit
A portable, multi-runtime engineering harness for AI coding agents — Claude Code · Codex · OpenCode · omp. Introspects your repo's stack and generates a harness tailored to it (a CLAUDE.md/AGENTS spine, a stack-specific architect agent, ver
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code $ claude mcp add example-db -- npx -y @example/mcp-server --readonly $ npx degit jhlee0409/omni-harness-kit/agents ./rig-omni-harness-kit/agents $ npx degit jhlee0409/omni-harness-kit/skills ./rig-omni-harness-kit/skills $ npx degit jhlee0409/omni-harness-kit/hooks ./rig-omni-harness-kit/hooks
MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ claude mcp add example-db -- npx -y @example/mcp-server --readonly $ npx degit jhlee0409/omni-harness-kit/adapters/omp/skills/accessibility-checks .claude/skills/accessibility-checks $ npx degit jhlee0409/omni-harness-kit/adapters/omp/skills/chrome-verify-checks .claude/skills/chrome-verify-checks $ npx degit jhlee0409/omni-harness-kit/adapters/omp/skills/db-verify-checks .claude/skills/db-verify-checks $ npx degit jhlee0409/omni-harness-kit/adapters/omp/skills/llm-eng-checks .claude/skills/llm-eng-checks $ npx degit jhlee0409/omni-harness-kit/adapters/omp/skills/perf-checks .claude/skills/perf-checks $ npx degit jhlee0409/omni-harness-kit/adapters/omp/skills/ui-verify-checks .claude/skills/ui-verify-checks $ npx degit jhlee0409/omni-harness-kit/skills/adr .claude/skills/adr $ npx degit jhlee0409/omni-harness-kit/skills/assess .claude/skills/assess $ npx degit jhlee0409/omni-harness-kit/skills/blast-radius .claude/skills/blast-radius $ npx degit jhlee0409/omni-harness-kit/skills/coding-guidelines .claude/skills/coding-guidelines $ npx degit jhlee0409/omni-harness-kit/skills/diagnose .claude/skills/diagnose $ npx degit jhlee0409/omni-harness-kit/skills/handoff .claude/skills/handoff $ npx degit jhlee0409/omni-harness-kit/skills/introspect .claude/skills/introspect $ npx degit jhlee0409/omni-harness-kit/skills/localize .claude/skills/localize $ npx degit jhlee0409/omni-harness-kit/skills/new-spec .claude/skills/new-spec $ npx degit jhlee0409/omni-harness-kit/skills/pickup .claude/skills/pickup $ npx degit jhlee0409/omni-harness-kit/skills/tdd .claude/skills/tdd $ npx degit jhlee0409/omni-harness-kit/skills/worktree .claude/skills/worktree
$ curl -fsSL --create-dirs -o .claude/agents/architecture-reviewer.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/architecture-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/change-verifier.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/change-verifier.md $ curl -fsSL --create-dirs -o .claude/agents/claim-checker.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/claim-checker.md $ curl -fsSL --create-dirs -o .claude/agents/instruction-critic.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/instruction-critic.md $ curl -fsSL --create-dirs -o .claude/agents/pr-shepherd.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/pr-shepherd.md $ curl -fsSL --create-dirs -o .claude/agents/readability-critic.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/readability-critic.md $ curl -fsSL --create-dirs -o .claude/agents/requirement-fidelity-critic.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/requirement-fidelity-critic.md $ curl -fsSL --create-dirs -o .claude/agents/spec-reviewer.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/spec-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/tdd-runner.md https://raw.githubusercontent.com/jhlee0409/omni-harness-kit/main/agents/tdd-runner.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} MCP servers (1)
| server | source | est. tokens |
|---|---|---|
| @example/mcp-server · "example-db" | npm | 2.5k |
Skills (18)
accessibility-checkschrome-verify-checksdb-verify-checksllm-eng-checksperf-checksui-verify-checksadrassessblast-radiuscoding-guidelinesdiagnosehandoffintrospectlocalizenew-specpickuptddworktree
Subagents (9)
| architecture-reviewer | >- |
| change-verifier | >- |
| claim-checker | >- |
| instruction-critic | >- |
| pr-shepherd | >- |
| readability-critic | >- |
| requirement-fidelity-critic | >- |
| spec-reviewer | >- |
| tdd-runner | >- |
Similar rigs
ooodnakov/ooodnakov-config
Opinionated eproducible personal dotfiles for Linux, Windows, and macOS machines.
Pragmatist 3.8k tok ·
johnstegeman/haven
—
Skill Collector 4.2k tok ·
daGrevis/Dotfiles
Dotfiles for my UNIX workstations
Automator 129 tok ·
kaisa-kucherenko/claude-code-flow
My working Claude Code setup: output styles, hooks, skills, agents
Orchestrator 1.1k tok ·