Nomos - Deterministic Git & AST Governance for AI Coding Agents
Server Details
Live GitHub repo AST X-Ray, 42 coding-agent DoD gates, Git worktree guards & brownfield baselines.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- mgantlett/nomos
- GitHub Stars
- 5
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 7 tools
Each tool targets a distinct intent—comparison, error remediation, gate catalog browsing, rollout planning, config scaffolding, handbook search, and repo auditing. Descriptions go further than most servers by including explicit 'Do NOT use when... use X instead' routing hints, eliminating overlap.
All seven tools follow a strict snake_case verb_noun pattern (compare_, diagnose_, inspect_, recommend_, scaffold_, search_, xray_) with clear object nouns. No style mixing or vague bare verbs.
Seven tools is well inside the sweet spot and each maps to a distinct phase of the governance workflow (evaluate, diagnose, browse, plan, scaffold, learn, audit). No filler or redundant entry points.
The advisory/read-only surface is well covered: comparison, error lookup, gate catalog, adoption planning, config templates, corpus search, and public-repo AST audit. Minor gaps exist—no private-repo or multi-repo batch audit path (only 'use local CLI' guidance) and no direct gate-by-severity or policy-diff operation, but agents can work around these.
Available Tools
7 toolscompare_governance_approachesCompare Nomos vs Linters, Hooks & Agent RulesARead-onlyIdempotentInspect
Compare Nomos deterministically against pre-commit hooks, SonarQube, ESLint, PR review bots (CodeRabbit, Greptile), prompt wrappers, and static CLAUDE.md / AGENTS.md / .cursorrules files. Example: focus='SonarQube' returns side-by-side trade-offs. Use when evaluating tool trade-offs or answering 'how is Nomos different from X'. Do NOT use when searching general Handbook tutorials (use search_nomos_corpus instead).
Inputs: focus. Returns: summary, canonicalUrl, dimensionCount
diagnose_nomos_errorInspect Nomos Gate or ERR_* Failure RemediationARead-onlyIdempotentInspect
Look up any Nomos ERR_* failure code, TC-AGT-* vector, or nomos verify gate rejection to retrieve its root cause and non-bypassing AST remediation. Example: error_code='ERR_CROSS_PLANE_MUTATION' returns the remediation guide. Use when troubleshooting a verify failure or pre-commit hook error. Do NOT use when browsing the full catalog of 42 gates by Law or category without an error (use inspect_dod_gates instead).
Inputs: error_log, error_code. Returns: query, docsUrl, summary, canonicalUrl, +2 more
inspect_dod_gatesInspect 42 Compiled DoD Gates & LawsARead-onlyIdempotentInspect
List and filter the 42 compiled Go and Tree-sitter AST Definition of Done quality gates, 7 Canonical Laws, and sandbox rules. Example: gate_id='G-16' or law=7 returns matching gate invariants. Use when browsing gate invariants by G-* ID, Law 1..7, or category. Do NOT use when troubleshooting a raw verify error log or ERR_* failure (use diagnose_nomos_error instead).
Inputs: law, gate_id, category. Returns: summary, totalGates, canonicalUrl, matchedCount
recommend_adoption_planRetrieve Progressive Repository Adoption PlanARead-onlyIdempotentInspect
Retrieve a read-only 4-step Nomos rollout guide (X-Ray baseline, stealth Branch Guard, CI SARIF, and isolated Git worktrees) tailored to repository LOC, languages, and air-gap constraints. Example: languages=['go','typescript'], repo_loc=120000 returns a phased brownfield guide. Use when planning rollout steps for a team or monorepo. Do NOT use when previewing raw nomos.yaml and CI workflow templates (use scaffold_nomos_config instead).
Inputs: repo_loc, airgapped, languages, ci_provider. Returns: summary, airgapped, stepCount, canonicalUrl
scaffold_nomos_configPreview nomos.yaml & CI Workflow TemplatesARead-onlyIdempotentInspect
Preview read-only nomos.yaml configuration text, 4-line GitHub Actions SARIF workflow text, and Day-0 onboarding commands without modifying files. Example: project_name='payments-api', mode='local' returns template text. Use when retrieving ready-to-paste YAML config and CI workflow templates. Do NOT use when requesting a strategic rollout recommendation by LOC or air-gap posture (use recommend_adoption_plan instead).
Inputs: mode, languages, ci_provider, project_name. Returns: mode, project, summary, nomosYaml, +4 more
search_nomos_corpusSearch Nomos Docs, Dispatches & SpecsARead-onlyIdempotentInspect
Search Nomos Engineering Handbook chapters, Empirical Field Dispatches, Open-Core Architecture Specs, and CLI reference by keyword. Example: query='worktree isolation', section='handbook' finds chapters on Git worktrees. Use when looking up conceptual architecture, empirical benchmarks, or CLI flags. Do NOT use when inspecting a specific G-* gate or ERR_* code (use inspect_dod_gates or diagnose_nomos_error instead) or comparing external tools (use compare_governance_approaches instead).
Inputs: query (required), section. Returns: query, summary, matchCount, canonicalUrl
xray_github_repoInspect Public GitHub Repo AST X-RayARead-onlyIdempotentInspect
Inspect a public GitHub repository via read-only static Tree-sitter AST analysis (<= 100MB source archive, strictly read-only) to retrieve its AI Hangover Score (0-100), complexity hotspots, and Day-0 baseline preview. Example: repo='go-chi/chi' returns the AST scorecard for go-chi/chi. Use when auditing a public GitHub owner/repo URL. Do NOT use when checking private repositories (private codebases require local CLI inspection via @mgantlett/nomos).
Inputs: repo (required). Returns: repo, grade, summary, canonicalUrl, +1 more
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
IaC, CI and dependency security review for coding agents: local scan CLI plus hosted OAuth MCP.
Screens public GitHub repos and PRs to generate risk maps, findings, and merge-readiness signals.
AI-native git hosting — repos, PRs, issues, CI gates, and AI code review over MCP (60 tools).
Code intelligence platform for AI agents. 20 tools for architecture, security & impact analysis.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceLive codebase intelligence for AI agents. Import graph PageRank for file importance, git forensics for co-change coupling and fragile code, convention detection across 16 domains, and blast radius analysis.16 npm3Business Source 1.1

VibeDoctorofficial
AlicenseNot gradedqualityBmaintenanceLocal repository health diagnosis and fix planning for AI-written JavaScript, TypeScript, Python and mixed projects. Its stdio MCP server exposes scans, finding explanations, repair plans, verification, privacy review and DPDP technical readiness checks.7 npm35GPL 3.0- AlicenseNot gradedqualityBmaintenanceIt enables AI coding agents to deterministically analyze TypeScript, React/Next.js, and Python code via AST parsing, tracing data flow, exposing hidden state leaks, and validating cross-boundary contracts with compact, token-efficient JSON. It also supports architecture rule auditing and drift detection between Git baselines and working trees.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI coding agents to query a local multi-repo code graph with token-budgeted context, impact analysis, bug-memory warnings, secret and prompt-injection scanning, snapshots, and git-worktree swarm orchestration. It also exposes telemetry resources for tracking token savings and security events.1MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.