Catch broken connections in agent-written code before it ships.
A mistyped environment variable. An import missing from the lockfile. A FastAPI route pointing at a handler that does not exist. Weftgate checks these connections against your repository through one local CLI, MCP server, hook, or GitHub Action.
No API key. No runtime dependencies. No network calls on the default verification path.
Watch the full demo · Transcript and reproducible evidence · How to write an oracle
pip install weftgate
cd /path/to/your/repo
weftgate doctor # see which contracts can be checked
weftgate audit # inspect the current repository
weftgate check app/main.py # check a file or directory
From source: pip install git+https://github.com/Avinash-Amudala/weftgate.git.
For development, bash scripts/bootstrap.sh installs the extras and runs the offline self-test.
The demo uses an intentionally broken fixture, not a field benchmark:
REJECT import 'requestz' → did you mean requests?
REJECT env var 'DATABSE_URL' → did you mean DATABASE_URL?
REJECT handler 'helth' → did you mean health?
Fix the three names and the fixture passes. A computed key such as os.environ[key]
returns review, because static analysis cannot establish its value.
To reproduce: weftgate eval mutate --fixture --seed 13.
| Result | Meaning | Blocks by default? |
|---|---|---|
accept |
The extracted reference resolves, or nothing checkable was found. | No |
review |
Evidence is incomplete or the reference is dynamic. | No |
reject |
A hard claim contradicts the indexed contract. | Yes |
unverifiable |
The relevant index or capability is unavailable. | No |
The rule is block only on a positive, machine-checkable falsehood. Suggestions
accompany findings when a nearby candidate exists. An accept verdict is not a test
suite result or proof that the application works; inspect coverage with weftgate doctor.
| Oracle | Checks | Conservative limits |
|---|---|---|
| Environment variables | Reads against dotenv examples, settings schemas, Docker/Compose declarations, code defaults, and configured files. | Dynamic keys and absent declaration sources soften. Declare externally supplied variables in your contract. |
| Python and Node imports | Imports against dependency metadata, local packages, known package aliases, and supported path aliases. | A manifest alone cannot prove a complete dependency tree. Unknown mappings and optional imports review. |
| FastAPI / Starlette routes | Handler references, router includes, and route claims using a static AST index. | Computed registration, external modules, and unresolved prefixes cannot be fully verified. |
This is an early static analyzer with bounded parsers. It does not execute your app, replace tests or a security scanner, or cover every framework. Python, Node, and monorepo import resolution can depend on runtime configuration. Missing or ambiguous evidence must soften; please report a false block with a minimal reproduction. Historical field runs document methods and limitations, rather than a claim of zero false positives.
weftgate setup --agents claude,cursor,vscode
weftgate setup --hooks # review the generated local configuration
The standard-library MCP server requires no extra package. Example MCP configuration:
{"mcpServers": {"weftgate": {"command": "weftgate", "args": ["mcp"]}}}
Run the server with the repository as its working directory. weftgate setup --agents all
writes supported project configs and prints snippets for clients with global settings.
CLI and MCP use the same gate functions. Tools include check_change, check_claim,
audit, suggest, index_status, and index.
git diff | weftgate check -
weftgate check --staged
weftgate check --path app/main.py --content proposed.py
weftgate claim '[{"kind":"route","method":"POST","path":"/users"}]'
weftgate audit --format=json
weftgate index --show routes
Exit codes: 0 allows the change, 1 blocks according to policy, 2 means a
usage or configuration error. Use --format=github for workflow annotations.
name: Verify connections
on: [pull_request]
permissions:
contents: read
jobs:
verify:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v5
with:
fetch-depth: 0
- uses: Avinash-Amudala/weftgate@v0.1.0
with:
mode: check # or audit to inspect the repository
The Action supports base, paths (shell-quoted paths, no shell expansion),
block-on, and python-version. Use a full checkout for diff ancestry.
The pre-commit hooks use the same gate; pin rev: v0.1.0.
[weftgate]
oracles = ["env_vars", "imports_lockfile", "routes_fastapi"]
block_on = "reject" # reject | review | never
env_declared_in = [".env.example"]
Precedence: WEFTGATE_* environment variables, repository configuration, user
configuration, then defaults. The index lives in the user cache, outside the repo.
weftgate doctor explains the selected configuration and missing contracts.
Outcome claims such as “tests passed” need machine-checkable evidence. Re-running a
named test command or probing a URL requires explicit --run and the configured
allowlist. See security and execution boundaries.
Weftgate exposes memory anchor, memory check, and memory changes through the CLI
and MCP. With mnemo installed in the same interpreter, add
"oracles": ["weftgate.memory"] to .mnemo.json to check structured memory claims.
Changed evidence keeps memories stale; checking does not silently replace the original hashes or prove remembered prose. See the integration contract. Mnemo is a separate companion repository and is not required to use Weftgate.
bash scripts/bootstrap.sh
.venv/bin/ruff check .
.venv/bin/ruff format --check .
.venv/bin/mypy weftgate
.venv/bin/python -m weftgate.selftest
.venv/bin/python -m pytest -q --cov=weftgate --cov-fail-under=85
Start with CONTRIBUTING.md, the oracle guide, and AGENTS.md. New oracles need a real broken example, a correct example, and a dynamic case that reviews. Reproducible false blocks and missing-contract reports are especially useful.
Apache-2.0 · Changelog · Launch plan