lhyman1701/gated-orchestration-template
A Claude Code project template that makes an AI agent build software under enforcement — deterministic machine gates, automatic test→requirement traceability, and an evidence-gated completion contract — instead of on model goodwill. Ships w
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 lhyman1701/gated-orchestration-template/.claude ./rig-gated-orchestration-template # inspect, then merge into .claude/ 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 lhyman1701/gated-orchestration-template/.claude/skills/entity-matching .claude/skills/entity-matching $ npx degit lhyman1701/gated-orchestration-template/.claude/skills/provenance-tracking .claude/skills/provenance-tracking $ npx degit lhyman1701/gated-orchestration-template/.claude/skills/rigor2 .claude/skills/rigor2 $ npx degit lhyman1701/gated-orchestration-template/.claude/skills/structured-parsing .claude/skills/structured-parsing $ npx degit lhyman1701/gated-orchestration-template/.claude/skills/traceability .claude/skills/traceability
$ curl -fsSL --create-dirs -o .claude/agents/auditor.md https://raw.githubusercontent.com/lhyman1701/gated-orchestration-template/master/.claude/agents/auditor.md $ curl -fsSL --create-dirs -o .claude/agents/build-executor.md https://raw.githubusercontent.com/lhyman1701/gated-orchestration-template/master/.claude/agents/build-executor.md $ curl -fsSL --create-dirs -o .claude/agents/kb-validator.md https://raw.githubusercontent.com/lhyman1701/gated-orchestration-template/master/.claude/agents/kb-validator.md $ curl -fsSL --create-dirs -o .claude/agents/research-executor.md https://raw.githubusercontent.com/lhyman1701/gated-orchestration-template/master/.claude/agents/research-executor.md $ curl -fsSL --create-dirs -o .claude/agents/traceability-checker.md https://raw.githubusercontent.com/lhyman1701/gated-orchestration-template/master/.claude/agents/traceability-checker.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"hooks": {
"PreToolUse": [
{
"matcher": "*",
"hooks": [
{
"type": "command",
"command": ".claude/hooks/cost_budget_abort.sh"
}
]
},
{
"matcher": "Write|Edit|MultiEdit",
"hooks": [
{
"type": "command",
"command": ".claude/skills/rigor2/hooks/rigor2-spec-gate.sh"
},
{
"type": "command",
"command": ".claude/skills/rigor2/hooks/rigor2-halt-gate.sh"
},
{
"type": "command",
"command": ".claude/hooks/pre_tool_use_block_frozen_dirs.sh"
},
{
"type": "command",
"command": ".claude/hooks/block_forbidden_keyword_in_settings.sh"
}
]
},
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": ".claude/skills/rigor2/hooks/rigor2-bash-secret-gate.sh"
}
]
},
{
"matcher": "Task",
"hooks": [
{
"type": "command",
"command": ".claude/skills/rigor2/hooks/rigor2-cove-isolation.sh"
}
]
},
{
"matcher": "WebFetch|WebSearch|mcp__fetch__fetch",
"hooks": [
{
"type": "command",
"command": ".claude/skills/rigor2/hooks/rigor2-source-quality.sh"
}
]
}
]
}
} Skills (5)
Subagents (5)
| auditor | Independent reviewer invoked at every phase boundary. Verifies code quality, test integrity, traceability completeness, and adherence to the project's hard requirements. Reports findings (PASSED / WAR |
| build-executor | Drives the build phase (prompts/02_build_prompt.md). Produces the application source and tests, reading the frozen knowledge directory as the source of truth. Reads STATE.md to resume; follows hard ta |
| kb-validator | Validates the knowledge directory's integrity. Run after the research phase completes, or any time the build phase hits an unexpected knowledge-base structure. Checks every cross-reference, every JSON |
| research-executor | Drives the research phase (prompts/01_research_prompt.md). Produces the project's knowledge directory (default kb/) of authoritative, cited domain knowledge. Reads STATE.md to resume; follows hard tas |
| traceability-checker | Checks the requirements traceability matrix for completeness and accuracy. Cross-references the matrix against the knowledge base and the actual test suite. Identifies orphaned requirements and orphan |
Hooks (16)
| event | matcher | runs |
|---|---|---|
| PreToolUse | * | .claude/hooks/cost_budget_abort.sh |
| PreToolUse | Write|Edit|MultiEdit | .claude/skills/rigor2/hooks/rigor2-spec-gate.sh |
| PreToolUse | Write|Edit|MultiEdit | .claude/skills/rigor2/hooks/rigor2-halt-gate.sh |
| PreToolUse | Write|Edit|MultiEdit | .claude/hooks/pre_tool_use_block_frozen_dirs.sh |
| PreToolUse | Write|Edit|MultiEdit | .claude/hooks/block_forbidden_keyword_in_settings.sh |
| PreToolUse | Bash | .claude/skills/rigor2/hooks/rigor2-bash-secret-gate.sh |
| PreToolUse | Task | .claude/skills/rigor2/hooks/rigor2-cove-isolation.sh |
| PreToolUse | WebFetch|WebSearch|mcp__fetch__fetch | .claude/skills/rigor2/hooks/rigor2-source-quality.sh |
| PostToolUse | Write|Edit|MultiEdit|Bash | .claude/skills/rigor2/hooks/rigor2-mcp-audit.sh |
| PostToolUse | Write|Edit|MultiEdit|Bash | .claude/skills/rigor2/hooks/rigor2-no-blockers.sh |
| PostToolUse | Write|Edit|MultiEdit|Bash | .claude/hooks/post_task_update_state.sh |
| SubagentStart | * | .claude/skills/rigor2/hooks/rigor2-context-quarantine.sh |
| Stop | * | .claude/skills/rigor2/hooks/rigor2-completion-gate.sh |
| Stop | * | .claude/skills/rigor2/hooks/rigor2-evidence-gate.sh |
| Stop | * | .claude/skills/rigor2/hooks/rigor2-verifier-independence.sh |
| Stop | * | .claude/skills/rigor2/hooks/rigor2-no-false-blocked.sh |
Slash commands (11)
/addfeature/analyze/audit/build/clarify/kb-check/kbgraph/research/resume/status/traceability
Similar rigs
carterlasalle/scc
System Context Compiler — compile code, config, infra, runtime, and intent into an evidence-backed machine model, then emit task-specific context packs for AI coding agents (Claude Code, Codex, Hermes, OpenCode).
Pragmatist 6.7k tok ·
SchotjeChrisman/agentic-workflow
A self-improving Claude Code setup
Orchestrator 2.8k tok ·
fotografvecerek-ai/claude-code-audit-gate
Independent audit agent for Claude Code: audits an app built by another agent (OWASP ASVS/Top 10 security, every feature, UI via Playwright, single source of truth, repo hygiene, git practice, token efficiency), never writes code, hands fin
Fort Knox 10.3k tok ·
homeofe/improvements
Portable, LLM-agnostic framework for multi-model AI workflows: model routing, structured agent handoff (AAHP), and a phased agent pipeline. MIT.
Orchestrator 1.4k tok ·