Creates, updates, or prunes an AGENTS.md for any repository by auditing the codebase, detecting non-discoverable gaps, and drafting minimal high-signal instructions that agents cannot infer from reading the code.
90
94%
Does it follow best practices?
Impact
78%
1.06xAverage score across 3 eval scenarios
Passed
No known issues
The Proxima engineering team has an existing AGENTS.md file that was written over a year ago when the project first started. Since then, the team has migrated from Pipenv to Poetry, introduced new linting and formatting tools, deprecated their v1 API endpoints in favor of v2, and added CI-specific database migration steps that differ from local workflows. The existing file has grown stale — it references outdated tools, contains sections that just describe the directory structure, and includes generic coding advice that doesn't help agents do anything specific.
A developer has asked you to bring the file up to date. The goal is not to preserve everything that's already there — it's to produce a tighter, more useful file. Anything that an agent could simply discover by reading the repository files should be removed. Anything stale (referencing removed tools or outdated commands) should be cut. Anything genuinely non-obvious and operationally important should be kept or added.
Produce the following files:
AGENTS.md — the updated agent instructions filechanges-summary.md — a record of what was removed from the original file and why, plus what was added and where it was foundThe following files are provided as inputs. Extract them before beginning.
=============== FILE: AGENTS.md ===============
This file contains guidance for AI agents working in this repository.
app/ — main application codeapp/api/ — API route handlersapp/models/ — SQLAlchemy modelsmigrations/ — Alembic migration filestests/ — test suiteInstall dependencies with pipenv:
pipenv install --devActivate the virtual environment:
pipenv shellThe /api/v1/ endpoints are deprecated and being migrated to /api/v2/. Do not add new features to /api/v1/ handlers. All new endpoint development should target /api/v2/.
pytest tests/=============== FILE: README.md ===============
A FastAPI service for the Proxima platform.
We use Poetry for dependency management.
poetry installRun the development server:
poetry run uvicorn app.main:app --reloadpoetry run pytest tests/See CONTRIBUTING.md for more details.
=============== FILE: pyproject.toml =============== [tool.poetry] name = "proxima-backend" version = "3.1.0" description = "Proxima platform backend"
[tool.poetry.dependencies] python = "^3.11" fastapi = "^0.109.0" sqlalchemy = "^2.0.25" alembic = "^1.13.1" redis = "^5.0.1"
[tool.poetry.dev-dependencies] pytest = "^7.4.4" ruff = "^0.2.0" mypy = "^1.8.0"
[tool.ruff] select = ["E", "F", "I"] line-length = 88
[tool.mypy] strict = true
=============== FILE: .pre-commit-config.yaml =============== repos:
=============== FILE: .github/workflows/ci.yml =============== name: CI
on: [push, pull_request]
jobs: test: runs-on: ubuntu-latest env: DATABASE_URL: postgresql://postgres:postgres@localhost:5432/proxima_test REDIS_URL: redis://localhost:6379 services: postgres: image: postgres:16 env: POSTGRES_PASSWORD: postgres ports: - 5432:5432 redis: image: redis:7 ports: - 6379:6379 steps: - uses: actions/checkout@v4 - uses: actions/setup-python@v5 with: python-version: '3.11' - run: pip install poetry && poetry install - run: poetry run alembic upgrade head --no-ansi - run: poetry run pytest tests/ --tb=short
=============== FILE: app/api/v1/users.py ===============
from fastapi import APIRouter router = APIRouter()
=============== FILE: app/api/v2/users.py =============== from fastapi import APIRouter router = APIRouter()
=============== FILE: CHANGELOG.md ===============