monte3l/m3l-groundwork
CLI that bootstraps a TypeScript project with a strict toolchain and Claude Code setup, or read-only surveys an existing one, plus a /customize plugin that tailors it to current guidance.
ARCHETYPE
Fort Knox
Deny lists, pre-tool hooks, sandboxing. Nothing touches prod without a signature.
GUARDRAILS
4/5
Blocks destructive commands · Protects secrets · Pre-tool screening hook · No YOLO mode · details
Quality 62/100 Guardrails v2
Copy this rig
# review before running: this installs third-party code
$ npx degit monte3l/m3l-groundwork/.claude ./rig-m3l-groundwork # 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 monte3l/m3l-groundwork/.claude/skills/creating-prs .claude/skills/creating-prs $ npx degit monte3l/m3l-groundwork/.claude/skills/customize .claude/skills/customize $ npx degit monte3l/m3l-groundwork/.claude/skills/finishing-work .claude/skills/finishing-work $ npx degit monte3l/m3l-groundwork/.claude/skills/grade-own-harness .claude/skills/grade-own-harness $ npx degit monte3l/m3l-groundwork/.claude/skills/harness-guidance .claude/skills/harness-guidance $ npx degit monte3l/m3l-groundwork/.claude/skills/starting-work .claude/skills/starting-work $ npx degit monte3l/m3l-groundwork/.claude/skills/triaging-ci .claude/skills/triaging-ci $ npx degit monte3l/m3l-groundwork/.claude/skills/typescript-guidance .claude/skills/typescript-guidance $ npx degit monte3l/m3l-groundwork/.claude/skills/writing-commits .claude/skills/writing-commits $ npx degit monte3l/m3l-groundwork/templates/packs/github/files/.claude/skills/reviewing-dependabot-prs .claude/skills/reviewing-dependabot-prs $ npx degit monte3l/m3l-groundwork/templates/packs/github/files/.claude/skills/triaging-scan-alerts .claude/skills/triaging-scan-alerts $ npx degit monte3l/m3l-groundwork/templates/packs/github/files/.claude/skills/watching-pr-checks .claude/skills/watching-pr-checks $ npx degit monte3l/m3l-groundwork/templates/packs/worktrees/files/.claude/skills/working-in-worktrees .claude/skills/working-in-worktrees
$ curl -fsSL --create-dirs -o .claude/agents/Explore.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/.claude/agents/Explore.md $ curl -fsSL --create-dirs -o .claude/agents/baseline-impact-reviewer.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/.claude/agents/baseline-impact-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/code-implementer.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/.claude/agents/code-implementer.md $ curl -fsSL --create-dirs -o .claude/agents/code-reviewer.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/.claude/agents/code-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/silent-failure-hunter.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/.claude/agents/silent-failure-hunter.md $ curl -fsSL --create-dirs -o .claude/agents/test-author.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/.claude/agents/test-author.md $ curl -fsSL --create-dirs -o .claude/agents/Explore.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/templates/core/.claude/agents/Explore.md $ curl -fsSL --create-dirs -o .claude/agents/code-implementer.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/templates/core/.claude/agents/code-implementer.md $ curl -fsSL --create-dirs -o .claude/agents/code-reviewer.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/templates/core/.claude/agents/code-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/silent-failure-hunter.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/templates/core/.claude/agents/silent-failure-hunter.md $ curl -fsSL --create-dirs -o .claude/agents/test-author.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/templates/core/.claude/agents/test-author.md $ curl -fsSL --create-dirs -o .claude/agents/type-design-analyzer.md https://raw.githubusercontent.com/monte3l/m3l-groundwork/main/templates/packs/quality/files/.claude/agents/type-design-analyzer.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(./secrets/**)",
"Read(./.env)",
"Read(./.env.*)",
"Read(./secrets/**)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-git-push-signed.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-double-background.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-hub-src-writes.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-git-push-signed.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-double-background.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-hub-src-writes.mjs\""
}
]
},
{
"matcher": "Write|Edit",
"hooks": [
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-js-extension.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-no-commonjs.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-protected-paths.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-branch-isolation.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-hub-src-writes.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-secret-writes.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-js-extension.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-no-commonjs.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-protected-paths.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-branch-isolation.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-hub-src-writes.mjs\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard-secret-writes.mjs\""
}
]
}
]
}
} Skills (13)
creating-prscustomizefinishing-workgrade-own-harnessharness-guidancestarting-worktriaging-citypescript-guidancewriting-commitsreviewing-dependabot-prstriaging-scan-alertswatching-pr-checksworking-in-worktrees
Subagents (12)
| Explore model: claude-haiku-5-5 | Fast read-only search agent for locating and understanding code. Use it to find files by pattern (e.g. "src/components/**/*.tsx"), grep for symbols or keywords (e.g. "API endpoints"), answer "where is |
| baseline-impact-reviewer model: claude-opus-5-5 | Read-only reviewer for changes under templates/core/ or templates/packs/. Checks the baseline's own contracts -- caps, domain-map coverage, dotfile escaping, pack budgets, "a pack never edits YAML/JS" |
| code-implementer model: claude-opus-5-5 | Writer spoke for the TDD build pipeline. Given a contract and a set of failing tests, writes the minimal src/** implementation to make those tests pass, then refactors while green. Use during the GREE |
| code-reviewer model: claude-opus-5-5 | Read-only reviewer for this project's source changes. Applies the four-part quality checklist and SOLID checks to a diff. Use after writing or changing source code, before commit. |
| silent-failure-hunter model: claude-opus-5-5 | Read-only error-handling auditor. Hunts for silent failures — swallowed exceptions, unchained causes, empty catch blocks, optional-chaining that masks errors, and retry/poll logic that exhausts withou |
| test-author model: claude-sonnet-5-5 | Writes Vitest tests for a source export — happy path, failure path, and expectTypeOf type-level tests where the type is the contract. This is the tests-first (RED) spoke of the TDD loop; it writes tes |
| Explore model: claude-haiku-5-5 | Fast read-only search agent for locating and understanding code. Use it to find files by pattern (e.g. "src/components/**/*.tsx"), grep for symbols or keywords (e.g. "API endpoints"), answer "where is |
| code-implementer model: claude-opus-5-5 | Writer spoke for the TDD build pipeline. Given a contract and a set of failing tests, writes the minimal src/** implementation to make those tests pass, then refactors while green. Use during the GREE |
| code-reviewer model: claude-opus-5-5 | Read-only reviewer for this project's source changes. Applies the four-part quality checklist and SOLID checks to a diff. Use after writing or changing source code, before commit. |
| silent-failure-hunter model: claude-opus-5-5 | Read-only error-handling auditor. Hunts for silent failures — swallowed exceptions, unchained causes, empty catch blocks, optional-chaining that masks errors, and retry/poll logic that exhausts withou |
| test-author model: claude-sonnet-5-5 | Writes Vitest tests for a source export — happy path, failure path, and expectTypeOf type-level tests where the type is the contract. This is the tests-first (RED) spoke of the TDD loop; it writes tes |
| type-design-analyzer model: claude-opus-5-5 | Read-only type-design reviewer. Rates the type design quality of changed exports on four dimensions (encapsulation, invariant expression, invariant usefulness, invariant enforcement), each scored 1–10 |
Hooks (38)
| event | matcher | runs |
|---|---|---|
| UserPromptSubmit | * | node "$CLAUDE_PROJECT_DIR/.claude/hooks/inject-decision-gate.mjs" |
| PreToolUse | Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-git-push-signed.mjs" |
| PreToolUse | Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-double-background.mjs" |
| PreToolUse | Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-hub-src-writes.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-js-extension.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-no-commonjs.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-protected-paths.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-branch-isolation.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-hub-src-writes.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-secret-writes.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/post-edit-verify.mjs" |
| PostToolUse | EnterWorktree|ExitWorktree|Agent | node "$CLAUDE_PROJECT_DIR/.claude/hooks/repair-core-bare.mjs" |
| SessionStart | startup|resume | node "$CLAUDE_PROJECT_DIR/.claude/hooks/repair-core-bare.mjs" |
| UserPromptSubmit | * | node "$CLAUDE_PROJECT_DIR/.claude/hooks/inject-decision-gate.mjs" |
| PreToolUse | Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-git-push-signed.mjs" |
| PreToolUse | Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-double-background.mjs" |
| PreToolUse | Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-hub-src-writes.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-js-extension.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-no-commonjs.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-protected-paths.mjs" |
| PreToolUse | Write|Edit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/guard-branch-isolation.mjs" |
Plugins (9)
skill-creator@claude-plugins-officialgithub@claude-plugins-officialsecurity-guidance@claude-plugins-officialtypescript-lsp@claude-plugins-officialclaude-code-setup@claude-plugins-officialclaude-md-management@claude-plugins-officialcontext7@claude-plugins-officialclaude-security@claude-plugins-officialplugin-dev@claude-plugins-official
Permissions
deny (6)
Read(./.env)
Read(./.env.*)
Read(./secrets/**)
Read(./.env)
Read(./.env.*)
Read(./secrets/**)
ask (0)
—
allow (0)
—
Similar rigs
aware-aeco/aware
Open-source agentic substrate for AECO. Apps are text. AI is the runtime.
Orchestrator 4.4k tok · · Q10
peterbamuhigire/chwezi-dev-engine
Software Dev't Engine for coding. Production-grade skills system for web apps & mobile architecture, APIs, databases, web, mobile, AI, security, UX, SaaS, and ERP engineering. Encodes reusable engineering judgment through layered baseline
Orchestrator 6.8k tok · · Q5
baksohyeon/mogui-ADE-orchestrator
Run one AI agent as the orchestrator of your whole workspace, inside the Orca ADE, on the CLI subscription you already pay for. No API key, no model endpoint. Orca hosts and places every session; the harness dispatches workers under contrac
Orchestrator 2.3k tok · · Q0
NYTC69/review-loop
Automates plan → implement → review iteration loops using multiple AI agents.
Orchestrator 3.7k tok · · Q0