Gnt

io.github.gnt-aiPublished by GitHub account gnt-aiv0.6.0Updated Nov 2, 2026

Git-native policy layer for AI agents: check_action verdicts against rules approved via PR.

VerifiedStreamable HTTPWeb executableKnowledge & MemoryDeveloper ToolsSecurity & Monitoring

Overview

AI-generated overview

Git-native policy layer that lets an assistant check a planned action against team-approved rules and get an allow, block or escalate verdict.

What you can do
  • Check a described action against approved rules and get an allowed, blocked or needs_human verdict with cited rules and a one-line reason
  • Search approved rules semantically, optionally filtered by tag
  • Fetch a single approved rule by id with its provenance (approver, date, citations)
  • List compiled skill pack versions for the org and fetch a pack's manifest and file list
When to use it
Use it when agents take consequential actions such as refunds, deletions or customer messages and you want each one gated by rules your team approves through merged pull requests, with a traceable paper trail.
Requirements
Remote MCP endpoint at api.gntai.dev. The companion CLI needs Node 22.13 or newer, a gnt login sign-in that stores an API key locally, and a connected GitHub repo for rules. Optional variables GNT_API_URL, GNT_WEB_URL and GNT_CONFIG_DIR change endpoints and config location.
Before you install
Rules are stored in the connected GitHub repo and in gnt's own database, and the MCP tools read from that store rather than your repo. Default rule extraction sends source text to Anthropic's API, and extracted rule text is sent to gnt's API to open pull requests. The CLI stores an API key in credentials.json. Verdicts fail closed to needs_human, so uncovered actions are not auto-approved.

Installation

In SourceWeft

  1. Open Gnt in the dashboard and add it to a workspace.
  2. 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/io.github.gnt-ai%2Fgnt/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.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "gnt": {
      "type": "http",
      "url": "https://api.gntai.dev/mcp"
    }
  }
}

README

Run the setup below once and every agent your team runs has a git-native rulebook it has to check over MCP before it acts, approved the same way your code already is: a merged pull request.

  • Rules live in your repo as files and ship through normal PRs, not a dashboard click.
  • Agents call check_action over MCP before anything risky (a refund, a delete, a message to a customer) and get back an allow/block/escalate verdict.
  • Every rule traces back to the git file and the PR that approved it, so "why did the agent do that" always has a paper trail.

Prefer not to run any of this yourself? The hosted version at gntai.dev does the same thing without you standing up a Postgres instance.

Get started (30 seconds)

bash
npm install -g @gnt-ai/clignt logingnt connect githubgnt prebrain# merge the opened PR on GitHub. that merge is the approval

Try it with Docker first

To see a real check_action response before installing Node or npm, clone the repository and run the Docker-only demo:

bash
./demo.sh

It builds the full local stack in an isolated gnt-demo Compose project, seeds an approved refund rule, calls check_action, and prints a curl command you can run again. Without an Anthropic key the real fail-closed path returns needs_human; pass GNT_DEMO_ANTHROPIC_API_KEY=sk-ant-... ./demo.sh to opt in to the model's grounded policy verdict. The demo uses deterministic local embeddings and throwaway localhost-only secrets, so it does not need a ZeroEntropy key.

[gnt connect's interactive picker, and what gnt prebrain's draft-PR output looks like]

Nothing for gnt prebrain to scan yet? Run gnt init first — it scaffolds rules/ locally with a couple of example rule files so there's something real to look at and edit, and points you at gnt prebrain --starter-packs for a curated pack to start from instead.

That merge lands a rule file in your connected repo, shaped like this:

your-repo/└── rules/    ├── refund-approval-threshold.md    └── contract-legal-cc.md

Each file is plain markdown with YAML frontmatter:

markdown
---title: Never refund over $500 without a managerstatus: approvedconfidence: 0.91owner_id: finance-teamsource_citations: [...]source: slacktags: [refunds, finance]last_validated_at: 2026-07-20version: 1superseded_by: nullapproved_by: [email protected]approved_at: 2026-07-21T14:03:00Zcreated_at: 2026-07-18T09:12:00Zpr_number: 142pr_url: https://github.com/your-org/your-repo/pull/142---
Refunds over $500 need manager sign-off before they go out...

See it in action

There's no captured transcript to show yet (see the gap noted at the bottom of this README). Here's the actual response shape a check_action call returns, straight from the tool's contract:

json
{  "verdict": "blocked",  "reason": "Refund exceeds the $500 threshold without manager sign-off (rules/refund-approval-threshold.md)",  "cited_rules": [    { "id": "refund-approval-threshold", "title": "Never refund over $500 without a manager" }  ],  "rules_retrieved": 3}

verdict is one of allowed, blocked, or needs_human. needs_human is the fail-closed default: no approved rule covers the action, retrieval failed, or the check couldn't complete. It never guesses.

What it does

One MCP endpoint, five tools:

ToolWhat it does
check_actionChecks a described action against your approved rules before an agent takes it. Returns allowed, blocked, or needs_human with cited rules and a one-line reason.
search_rulesSemantic search over your org's approved rules, optionally filtered by tag. An empty list means no approved rule covers the query.
get_ruleFetches one approved rule by id, with its provenance (who approved it, when, what it was cited from).
list_skill_packsLists every compiled skill pack version for your org, newest first.
get_skill_packFetches a compiled skill pack's manifest and file list by id.

Prerequisites

RequirementCheckGet it
Node >=22.13node --versionnodejs.org

Install

MethodCommand
curlcurl -fsSL gntai.dev/install.sh | sh
npmnpm install -g @gnt-ai/cli

gnt needs Node >=22.13. If the CLI fails to start with a version error, update Node first and confirm with node --version.

Common commands

bash
gnt login                # sign in, store an API key locallygnt init                 # scaffold a local rules/ dir with example rule filesgnt connect github       # connect the repo your rules PRs open againstgnt prebrain             # scan sources, extract candidate rules, open PRsgnt review               # review rules awaiting approvalgnt status               # show brain statusgnt pull                 # download the latest skill packgnt gaps                 # list uncovered queries with no approved rule

Config

VariableDefaultWhat it controls
GNT_API_URLhttps://api.gntai.devAPI endpoint the CLI and MCP calls hit
GNT_WEB_URLhttps://gntai.devWeb app used for gnt login's browser step
GNT_CONFIG_DIR~/.gntWhere credentials.json and local config live

Privacy

  • No analytics or telemetry dependency in the CLI or the web app.
  • gnt prebrain's default extraction mode is cloud, not on-device: your source text goes straight to Anthropic's API (or Vercel AI Gateway with zero-data-retention, if you configure it), never to gnt's own servers. Fully on-device extraction needs --mode local against a local Ollama daemon.
  • The extracted rule candidates still get sent to gnt's API to open the PR. Raw source text stays off gnt's servers in cloud mode; the resulting rule text doesn't.
  • Rules live in your connected GitHub repo and in gnt's own database. The MCP tools read from gnt's store, not by cloning your repo on every call.
  • Self-hosting: apps/api only sends error data to Sentry if you set SENTRY_DSN yourself. Leave it unset and nothing goes out.

Team setup

  • Who writes rules: anyone with access to your connected repo, either through gnt prebrain (batch-extracted from real sources) or gnt review (hand-proposed).
  • How approval works: merging the PR is the approval. There's no separate publish step.
  • What gets committed: rules/<rule-id>.md files with the frontmatter shown above and a plain markdown body.
  • Catching a malformed rule before it's reviewed: gnt rules lint checks a rule file's frontmatter locally, and gnt-ai/gnt/.github/actions/lint-rules runs the same check as a CI step on your rules repo's own PRs, so a bad frontmatter fails the PR instead of the review.

Troubleshooting

Self-hosting: gnt login's browser step has nowhere to land. gnt login opens a browser to a /cli-login page and polls the API for the resulting key — that page is served by the hosted product's web app, which isn't part of this repo. There's no CLI-only login flow (device code or otherwise) today, and no gnt command to set a key manually. Self-hosting this stack currently means building your own thin frontend for that one route (it just needs to complete the sign-in flow and hand the CLI a key). This is a real, open gap in the self-host path, not a config issue — closing it properly means adding a CLI-only login flow.

ValueError: refusing to start: these settings still have their .env.example placeholder value... A change-me-... string is still sitting in apps/api/.env. The error names every offending field; generate a real value for each and retry.

store fails to start with GNT_STORE_INTERNAL_API_SECRET is not set. apps/store/.env wasn't filled in, or wasn't picked up. Confirm the file exists at that exact path, not still named .env.example.

Every store-to-api call gets rejected with 401 or 403, even though both services are up. STORE_INTERNAL_API_SECRET / APPROVAL_SIGNING_SECRET in apps/api/.env don't byte-for-byte match GNT_STORE_INTERNAL_API_SECRET / GNT_APPROVAL_SIGNING_SECRET in apps/store/.env. This fails closed by design. Regenerate both pairs so the two files agree.

A rule fails to save with an embedding or rerank error. apps/store/.env is missing ZEROENTROPY_API_KEY, or it's still empty. Get a real one from zeroentropy.dev.

Full command reference

gnt logingnt logoutgnt init                 scaffold a local rules/ dir with example rule files (--dir <path>)gnt connect <app>        github, slack, notion-mcp, monday-mcp, linear-mcp, jira-mcp,                          sentry-mcp, granola-mcp, zoom-mcp, figma, datadog,                          gitlab-threads, hubspot, airtable, openclaw, hermesgnt disconnect <app>gnt statusgnt billinggnt reviewgnt pullgnt gapsgnt prebrain              scan local sources, extract candidate rules, open batched draft                           PRs (~60 flags for source paths and extraction mode, see                           `gnt prebrain --help`; --mode cloud|local, cloud is the default)gnt stalegnt keys list|create|revoke|rotategnt webhook list|create|revokegnt org show|rename|invite|remove

Shell completion

bash
# bash, add to ~/.bashrceval "$(gnt completion bash)"
# zsh, add to ~/.zshrceval "$(gnt completion zsh)"
# fish, add to ~/.config/fish/config.fishgnt completion fish | source

Learn more

Self-hosting is a first-class, fully supported path — Apache-2.0 from day one, run it on your own infra with your own keys, or use the hosted version at gntai.dev. The homepage FAQ and the self-hosting docs both describe this same path; there is no "not today" caveat.

License

Copyright © 2026 gnt.ai. Licensed under Apache-2.0 — see LICENSE for the terms and NOTICE for the trademark rule on forks.

Why is this free?

Self-hosting gnt costs you nothing, forever — clone it, run docker compose up, bring your own keys. What we sell is the part self-hosting doesn't give you: hosting at gntai.dev, managed OAuth connectors (GitHub, Slack, Linear, Notion, Zendesk — no app-approval process on your end), and usage-based AI features. If you'd rather run it yourself, that's a fully supported, fully free path, not a crippled trial of the real thing.

Contributing

See CONTRIBUTING.md for dev setup and how to open a PR. Every commit needs a Signed-off-by trailer (git commit -s), the Developer Certificate of Origin instead of a CLA. No separate form, just the flag.

Thanks to everyone who's sent a PR:

[gnt contributors]

Source: README.md at commit 59a743d

Tools

0
SourceWeft has not read this server's tools yet.

Version history

1
  1. v0.6.0LatestSep 16, 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