
Nomos - Deterministic Git & AST Governance for AI Coding Agents
ca.sophialabsPublished by sophialabs.caOfficial publisherv4.30.0Updated Nov 22, 2026
Live GitHub repo AST X-Ray, 42 coding-agent DoD gates, Git worktree guards & brownfield baselines.
Overview
Nomos - Deterministic Git & AST Governance for AI Coding Agents's hosted MCP server (sophialabs.ca), published by sophialabs.ca; no sign-in required; Streamable HTTP.
No AI overview yet: these facts come from the registry entry and the server itself.
- Publisher
- sophialabs.ca
- Endpoint
- sophialabs.ca
- Last checked
Installation
In SourceWeft
- Open Nomos - Deterministic Git & AST Governance for AI Coding Agents in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/ca.sophialabs%2Fnomos/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config.
{
"mcpServers": {
"nomos-deterministic-git-ast-governance-f": {
"type": "http",
"url": "https://sophialabs.ca/mcp"
}
}
}README
Nomos (Νόμος)
Agents guess; Git cannot. The Apache 2.0 Go execution substrate beneath any coding agent (0 LLM tokens, certified up to 3.60M LOC).
[CI] [Release] [License] [Docs] [Nomos X-Ray] [Telemetry]
Nomos is a single ~20 MB static Go binary (Apache 2.0, 0 LLM tokens, zero cloud account) between your coding agent (Claude Code, Cursor, OpenCode, Codex, Gemini CLI, Antigravity, Windsurf, or Aider) and Git. Run nomos init once (or paste 1 prompt into your agent) - Nomos blocks skipped tests, hollow stubs, monolith bloat, and leaked secrets in 24ms (<365ms full 42-gate diff verify even across 3.60M-LOC monorepos), your agent fixes its own mess mid-turn, and your feat/* workflow never changes:
Initialize once, then forget
nomosexists. Nomos supplies0-token repository physics; your coding agent supplies the intelligence to self-correct before ending its turn. Usenomos xray .anytime you want a~4scodebase health scan.
Two Ways to Set Up in 30 Seconds (Or Scan in Browser First)
Zero-Install Browser Scan (5s): Paste any public GitHub repository at sophialabs.ca/xray - or just replace
github.comwithsophialabs.cain your browser URL bar (e.g.https://sophialabs.ca/go-chi/chi) - to inspect its 11-languageAI Hangover Score(0-100) and download a pre-computednomos.baseline.json+ 4-line GitHub Action PR gate before installing anything locally.
Path A - Paste 1 Prompt Into Your Coding Agent (Zero Manual Setup):
Path B - Run 3 Commands in Your Terminal:
What happens automatically underneath your agent the next time it tries taking a shortcut on feat/*:
Your coding agent receives that deterministic AST fix recipe mid-turn and repairs billing.go itself before handing control back to you.
How It Works Without Changing Your Workflow (100% Apache 2.0 Open-Core)
Every single-repository governance capability ships locally in the open-source nomos binary with zero crippled commands, organized around three clear boundaries:
1. What You Touch Once (30 Seconds: nomos xray, nomos init)
- On-Demand 5s Stethoscope (
nomos xray . [--markdown]): Walksgit ls-filesacross 11 languages for anAI Hangover Score(0-100,Grade A-F), clones, monoliths (>400 LOC), stubs, swallowed errors, secret/host-path leaks, CWE sinks, and ready-to-lockarchitecture_boundarieswith zero repository writes. Run it on Day 0 or whenever you want a codebase report card. - Non-Breaking Legacy Drop-In (
nomos init): Automatically freezes Day-0 debt bounds innomos.baseline.json(strict_mode: falseon brownfield, strict anywhere via--strict) so zero existing legacy files fail, ratcheting only touched files (<= 0added complexity/debt,>= 0coverage delta,0new clones or swallowed errors; self-hostednomosholds0/100 Grade Aunderstrict_mode: true). Auto-detects installed agents, scaffolds a compact~13-lineAGENTS.mdBranch Guard block, and wires native lifecycle + Git hooks (nomos --helpshows everyday commands;nomos --help-agentkeeps internal agent RPCs out of your way). - Zero-Footprint Stealth Trial (
nomos init --local): Registers generated configs and hooks in.git/info/excludeinstead of.gitignore, leavinggit status100% clean while you evaluate solo; promote anytime withnomos init.
2. What Runs Invisibly Underneath Every Agent Turn (0 Commands to Learn)
- Protected
main& Automaticfeat/*Verification: Defaultisolation_mode: "branch"blocks directmain/masterwrites and config/hook tampering (nomos.yaml,nomos.baseline.json,.git/hooks/*), while plaingit commitonfeat/*branches runs42polyglot AST gates across Go, TypeScript/Svelte, Python, Rust, Ruby, C/C++, Java, Kotlin, and C# in~460ms(~24mssingle-file radar,0LLM tokens) - catching skipped/deleted tests (t.Skip,@pytest.mark.skip,#[ignore],xit,@Disabled,[Fact(Skip=...)]), hollow stubs (todo!(),pass,NotImplementedError),>60 LOCfunctions (400 LOCfile ceiling),architecture_boundaries, andstrict_types. ~4msPre-Tool Security Airlock & Loop Breaker (nomos hook): Denies agent reads of~/.ssh,~/.aws,~/.kube,~/.gnupg,~/.config/gcloud,.netrc,.npmrc,.pypirc, blocks cloud metadata SSRF (169.254.169.254), rejects unverified "tests pass" claims, and halts 3 consecutive non-progressing tool failures before burning context windows.- Verified-Only Episodic Memory (
mneme.db/nomos mcp):.git/hooks/post-commitdistillstrap -> fixlessons into pure-Go SQLite (mneme.db,<50mslexical/vector recall + read-onlystdioMCP vianomos mcp) only after verification passes, so broken attempts never poison future agent sessions.
3. Optional Scale-Up When You Outgrow Single Branches (--ci, --worktrees)
- Zero-Trust CI Diff-Mutation Proofs (
nomos init --ci,nomos verify --pr --mutate-diff): Mutates newly added conditionals/returns across Go, TS/JS, and Python so tautological tests fail closed on PRs, emits OASIS SARIF 2.1.0 for GitHub Code Scanning, and posts PR receipts viaaction.yml. - Parallel Multi-Agent Worktrees (
nomos init --worktrees): Upgrade on demand when running multiple concurrent agents to enforce disposablegit worktreesandboxes, live exported-symbol collision radar (MTPO), and causal spec-before-code (INT/SUB) separation (see Architecture Blueprint below).
Install Alternatives & Graduation Recipes
Alternative Package Managers & Updates
Update or cleanly uninstall anytime:
Optional Graduation: CI PR Proofs & Parallel Worktrees
Install the official Nomos Zero-Trust PR Proof GitHub Action from GitHub Marketplace, or scaffold CI workflows locally when ready:
Enforce zero-trust PR mutation proofs in GitHub Actions (GitHub Marketplace):
Cheap Workhorse Stack (
OpenCode+OpenRouter/ Local16GBGPU): Because.opencode/plugins/nomos.tsenforces frontier merge rules in local compiled Go (0middleman token tax),14B-32Bopen-weight models jump from27.8%to86.1%brownfield resolution (3.1xlift) while cutting token spend-43%to-57%.
Why Nomos Is a New Layer of the Stack
Verified Benchmarks & Runtime Footprint
Reproduce locally in one command:
go -C substrate test ./cli -run "TestTraceTrialBench|TestCleanroomPolyglotObligation|TestCleanroomDownstreamVectors" -v
Expand Full Per-Model Benchmark Tables (108-Run GPU Matrix, 1,541-Run Security/OSS Audit, Harbor & TraceTrial-Bench)
1. Open-Weight & Commodity Workhorse Brownfield Lift (N = 108 Paired Episodes · 8B to 32B)
Across 108 paired zero-hint brownfield episodes (54 Stock vs. 54 Nomos v4.13.1, temp=0.0) spanning Python, TypeScript, and Go on 6 open-weight models sized for single-card 16GB-24GB GPUs (RTX 4080, RTX 3090, A10G, L4):
2. Zero-Trust Security & Real-Repo OSS Debt Containment (1,587 Live Trajectories)
3. Terminal-Bench 2.0 Harbor Hub Trajectories & TraceTrial-Bench (arXiv:2610.32285)
Terminal-Bench 2.0(89 / 89Tasks viaHarbor·nomos-harbor): Multi-trialk=5evaluation (445Docker trials onGemini Flash) achieved76.40%Pass@1(340/445) and92.13%Pass@5(82/89tasks solved) at91.20%prompt cache stability. In paired single-shotGemini CLI 0.67.0A/B runs, Stock scored58 / 89(65.17%,49f21995) vs.63 / 89(70.79%,+5.62 pp) single-shot (39059e9a) and75 / 89(84.27%,+19.10 pp)pass<=2(6b9f81a5).- Frontier Harness Fault-Containment (
TraceTrial-Bench·arXiv:2610.32285): Across190paired Docker runs (95tasks holdingClaude Code v2.1.220+Gemini 2.5 Flash-Litefixed),nomos hooklifted Context & Continuity by+78.3%, cutharness_causedfailures by-18.2%(55->45), and raised reliability from44.89to48.78.
Expand Architecture Blueprint, Worktrees (`INT`/`SUB`), CLI Surface & Specifications
When your team scales beyond single-branch workflows (nomos init --worktrees):
- Hermetic Worktree Isolation: Direct edits in the root checkout fail closed;
nomos task chore/nomos task startmintscapbac.Leasecapabilities inside disposableworktrees/<repo>-<task>sandboxes (PLAN -> EDIT -> REVIEW -> SYNC -> LEARN). - Two-Plane Spec-vs-Code Separation (
INT/SUB) &2PCSync: Separates human specifications (INT-*) from substrate implementation (SUB-*), whilenomos task syncruns42gates, fast-forwardsmain, records a SHA-256 Git Notes receipt (refs/notes/nomos/releases), and prunes the worktree. - Multi-Worktree AST Collision Radar (
MTPO) &NCTE-100:nomos physicscomputes live exported symbol hulls across active sandboxes to prevent merge collisions, andnomos directive --lintenforces unambiguous controlled English inintent/specs.
- Scope of Proof: Nomos verifies
<5mscompiled structural and traceability invariants (import boundaries, contract field parity, anti-shim AST checks,TC-*test bindings, and SHA-256 spec digests) rather than LLM-as-a-Judge critics or formal theorem provers (Coq,Lean,TLA+); runtime semantics remain governed by your test suite.
Expand The 7 Laws of Nomos (42-Gate Invariant Matrix)
Full 42-gate matrix in nomos_intent_traceability_spec.md:
Open-Core Boundary
Nomos Community Edition is 100% open-source under Apache 2.0 with zero telemetry, zero cloud account requirements, and zero crippled CLI commands. Every single-developer and single-repository governance capability runs locally out of the box (see LICENSE, NOTICE, and intent/intent.md):
Community & License
- Community & Governance: Join GitHub Discussions for setup Q&A,
nomos.yamltuning, and feature ideas. SeeCONTRIBUTING.md,CODE_OF_CONDUCT.md, andSECURITY.md. - License & Citation: Licensed under the Apache License, Version 2.0 (see
LICENSE,NOTICE, andTHIRD_PARTY_NOTICES.md). For academic citation, seeCITATION.cffand DOI10.5281/zenodo.24240539.
Built with deterministic rigor by SophiaLabs · Governed by Nomos
Source: README.md at commit db330ce
Tools
0Version history
1- v4.30.0LatestNov 22, 2026
Score
85% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 daysRequired25 pts
- Auth declaredRequired10 pts
- Tools listedRequired15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
