sd0xdev/sd0x-harness
The harness layer for Claude Code — a reference implementation of harness engineering with hook-enforced dual review, state-machine gates that survive context compaction, and fail-closed safety where it counts. Quality gates that AI can't s
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code $ npx degit sd0xdev/sd0x-harness/agents ./rig-sd0x-harness/agents $ npx degit sd0xdev/sd0x-harness/skills ./rig-sd0x-harness/skills $ npx degit sd0xdev/sd0x-harness/hooks ./rig-sd0x-harness/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.
$ npx degit sd0xdev/sd0x-harness/skills/adr .claude/skills/adr $ npx degit sd0xdev/sd0x-harness/skills/agentctl-setup .claude/skills/agentctl-setup $ npx degit sd0xdev/sd0x-harness/skills/architecture .claude/skills/architecture $ npx degit sd0xdev/sd0x-harness/skills/ask .claude/skills/ask $ npx degit sd0xdev/sd0x-harness/skills/best-practices .claude/skills/best-practices $ npx degit sd0xdev/sd0x-harness/skills/bug-fix .claude/skills/bug-fix $ npx degit sd0xdev/sd0x-harness/skills/bump-version .claude/skills/bump-version $ npx degit sd0xdev/sd0x-harness/skills/check-coverage .claude/skills/check-coverage $ npx degit sd0xdev/sd0x-harness/skills/claude-health .claude/skills/claude-health $ npx degit sd0xdev/sd0x-harness/skills/code-explore .claude/skills/code-explore $ npx degit sd0xdev/sd0x-harness/skills/code-investigate .claude/skills/code-investigate $ npx degit sd0xdev/sd0x-harness/skills/codex-architect .claude/skills/codex-architect $ npx degit sd0xdev/sd0x-harness/skills/codex-brainstorm .claude/skills/codex-brainstorm $ npx degit sd0xdev/sd0x-harness/skills/codex-cli-review .claude/skills/codex-cli-review $ npx degit sd0xdev/sd0x-harness/skills/codex-code-review .claude/skills/codex-code-review $ npx degit sd0xdev/sd0x-harness/skills/codex-explain .claude/skills/codex-explain $ npx degit sd0xdev/sd0x-harness/skills/codex-implement .claude/skills/codex-implement $ npx degit sd0xdev/sd0x-harness/skills/codex-review-branch .claude/skills/codex-review-branch $ npx degit sd0xdev/sd0x-harness/skills/codex-review-doc .claude/skills/codex-review-doc $ npx degit sd0xdev/sd0x-harness/skills/codex-review-fast .claude/skills/codex-review-fast $ npx degit sd0xdev/sd0x-harness/skills/codex-review .claude/skills/codex-review $ npx degit sd0xdev/sd0x-harness/skills/codex-security .claude/skills/codex-security $ npx degit sd0xdev/sd0x-harness/skills/codex-setup .claude/skills/codex-setup $ npx degit sd0xdev/sd0x-harness/skills/codex-test-gen .claude/skills/codex-test-gen $ npx degit sd0xdev/sd0x-harness/skills/codex-test-review .claude/skills/codex-test-review $ npx degit sd0xdev/sd0x-harness/skills/contract-decode .claude/skills/contract-decode $ npx degit sd0xdev/sd0x-harness/skills/create-pr .claude/skills/create-pr $ npx degit sd0xdev/sd0x-harness/skills/create-request .claude/skills/create-request $ npx degit sd0xdev/sd0x-harness/skills/de-ai-flavor .claude/skills/de-ai-flavor $ npx degit sd0xdev/sd0x-harness/skills/debug .claude/skills/debug $ npx degit sd0xdev/sd0x-harness/skills/deep-analyze .claude/skills/deep-analyze $ npx degit sd0xdev/sd0x-harness/skills/deep-explore .claude/skills/deep-explore $ npx degit sd0xdev/sd0x-harness/skills/deep-research .claude/skills/deep-research $ npx degit sd0xdev/sd0x-harness/skills/dep-audit .claude/skills/dep-audit $ npx degit sd0xdev/sd0x-harness/skills/deploy-flow .claude/skills/deploy-flow $ npx degit sd0xdev/sd0x-harness/skills/dev-security-audit .claude/skills/dev-security-audit $ npx degit sd0xdev/sd0x-harness/skills/doc-refactor .claude/skills/doc-refactor $ npx degit sd0xdev/sd0x-harness/skills/doc-review .claude/skills/doc-review $ npx degit sd0xdev/sd0x-harness/skills/epic-merge .claude/skills/epic-merge $ npx degit sd0xdev/sd0x-harness/skills/feasibility-study .claude/skills/feasibility-study
$ curl -fsSL --create-dirs -o .claude/agents/architecture-designer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/architecture-designer.md $ curl -fsSL --create-dirs -o .claude/agents/brief-writer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/brief-writer.md $ curl -fsSL --create-dirs -o .claude/agents/code-simplifier.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/code-simplifier.md $ curl -fsSL --create-dirs -o .claude/agents/codex-architect.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/codex-architect.md $ curl -fsSL --create-dirs -o .claude/agents/codex-implementer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/codex-implementer.md $ curl -fsSL --create-dirs -o .claude/agents/contract-neutral-reviewer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/contract-neutral-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/coverage-analyst.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/coverage-analyst.md $ curl -fsSL --create-dirs -o .claude/agents/doc-refactor.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/doc-refactor.md $ curl -fsSL --create-dirs -o .claude/agents/feasibility-analyst.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/feasibility-analyst.md $ curl -fsSL --create-dirs -o .claude/agents/git-investigator.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/git-investigator.md $ curl -fsSL --create-dirs -o .claude/agents/performance-optimizer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/performance-optimizer.md $ curl -fsSL --create-dirs -o .claude/agents/refactor-reviewer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/refactor-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/solution-architect.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/solution-architect.md $ curl -fsSL --create-dirs -o .claude/agents/strict-reviewer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/strict-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/tech-spec-reviewer.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/tech-spec-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/verify-app.md https://raw.githubusercontent.com/sd0xdev/sd0x-harness/main/agents/verify-app.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 6,974 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(~/.aws/**)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(.env)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/.env.*)",
"Read(**/*.key)",
"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(npm publish:*)",
"Bash(wget *)",
"Bash(git rebase *)",
"Bash(gh pr merge *)",
"Bash(git commit *)"
]
}
} Skills (102)
adragentctl-setuparchitectureaskbest-practicesbug-fixbump-versioncheck-coverageclaude-healthcode-explorecode-investigatecodex-architectcodex-brainstormcodex-cli-reviewcodex-code-reviewcodex-explaincodex-implementcodex-review-branchcodex-review-doccodex-review-fastcodex-reviewcodex-securitycodex-setupcodex-test-gencodex-test-reviewcontract-decodecreate-prcreate-requestde-ai-flavordebugdeep-analyzedeep-exploredeep-researchdep-auditdeploy-flowdev-security-auditdoc-refactordoc-reviewepic-mergefeasibility-studyfeature-devfeature-verifyfp-briefgenerate-runnergh-stackgit-investigategit-profileinstall-hooksinstall-rulesinstall-scriptsissue-analyzejiraload-pr-reviewmerge-prepnecessity-auditnext-stepobsidian-cliop-sessionorchestrateplan-reviewportfoliopost-dev-recappost-dev-testpr-commentpr-reviewpr-summarypre-pr-auditprecommit-fastprecommitproject-auditproject-briefproject-setuppush-cirecap-askrecap-docrefactorremindrepo-intakereq-analyzerequest-trackingreview-specrisk-assessrunbooksafe-removesecurity-reviewseek-verdictsharingansimplifyskill-health-checksmart-commitsmart-rebasestatusline-configtech-brieftech-spectest-deeptest-healthtest-reviewui-first-principlesupdate-docsverifywatch-cizh-tw
Subagents (16)
| architecture-designer model: opus | Architecture design expert. Synthesizes code analysis, tech-spec context, and architecture advice into structured architecture documents with component diagrams, data flows, and architecture decisions |
| brief-writer model: opus | PM/CTO executive summary expert. Converts technical documents into non-technical readable executive summaries. |
| code-simplifier model: opus | Cleanup refactoring expert. Simplifies code, eliminates duplication, preserves behavior. |
| codex-architect model: opus | Codex architecture consulting expert (third brain). Uses Codex CLI to obtain architecture advice and integrates dual perspectives. |
| codex-implementer model: opus | Codex implementation expert. Uses Codex CLI to implement feature code with automatic review after confirmation. |
| contract-neutral-reviewer model: opus | Contract-neutral fallback reviewer. Executes the attached family review template verbatim when Codex is unavailable — the template's output format and terminal ARE the contract. Independent research, |
| coverage-analyst model: opus | Test coverage analysis expert. Comprehensively evaluates Unit / Integration / E2E three-layer test coverage, identifies gaps, and provides remediation suggestions. |
| doc-refactor model: opus | Document refactoring expert. Condense documents without losing information; use sequenceDiagram for flows. |
| feasibility-analyst model: opus | Feasibility analysis expert. Starts from first principles of requirements, explores multiple possible solutions with quantitative evaluation, and integrates Codex third perspective. |
| git-investigator model: opus | Code archaeology expert. Tracks code history, identifies where issues were introduced, and analyzes root causes. |
| performance-optimizer model: opus | Performance optimization expert. Identifies N+1 queries, memory leaks, and slow queries. |
| refactor-reviewer model: opus | Refactoring review expert. Evaluates refactoring risk and verifies behavior preservation. |
| solution-architect model: opus | Solution deepening expert. Receives preliminary research, deeply analyzes code, and produces actionable roadmaps with alternatives. |
| strict-reviewer model: opus | Strict code reviewer. Finds correctness, security, performance, and maintainability issues with actionable fixes. Use proactively after code changes. |
| tech-spec-reviewer model: opus | Technical spec review expert. Reviews technical documents for completeness, feasibility, and risk assessment, and provides improvement suggestions. |
| verify-app model: opus | Verification expert. Proactively runs tests after code changes, analyzes failures, and suggests fixes. |
Similar rigs
AsiaOstrich/universal-dev-standards
Universal, language-agnostic development standards for software projects. Includes coding standards, git workflows, testing guidelines, documentation structure, and AI collaboration rules.
Skill Collector 13.3k tok ·
anujg21/claude-engineering-os
A repository template that gives Claude Code a working engineering method: phased workflow, path-scoped standards, independent review agents, and hooks that block dangerous commands.
Skill Collector 3.1k tok ·
Yeachan-Heo/oh-my-claudecode
Teams-first Multi-agent orchestration for Claude Code
Orchestrator 14.5k tok ·
oprogramadorreal/optimus-claude
Primes your project for peak Claude Code performance
YOLO Cowboy 1.5k tok ·