Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/jettyio/mlcbakery/claude-mdgit clone --depth 1 https://github.com/jettyio/mlcbakeryWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01339 | $0.01339 |
| Opus 5 | $0.00669 | $0.00669 |
| Sonnet 5 | $0.00268 | $0.00268 |
| Haiku 4.5 | $0.00134 | $0.00134 |
Grade C, and why
mlcbakery CLAUDE.md scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -LsSf https://astral.sh/uv/install.sh | sh Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -LsSf https://astral.sh/uv/install.sh | sh How it starts
The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
MLC Bakery is a machine learning data and model management system with provenance tracking, built with FastAPI and SQLAlchemy. It provides RESTful APIs for managing ML model lineage, datasets, collections, and activities with support for Croissant metadata validation.
Architecture
The application follows a layered architecture:
- API Layer (
mlcbakery/api/) - FastAPI routers and endpoints organized by resource type - Schema Layer (
mlcbakery/schemas/) - Pydantic models for request/response validation - Model Layer (
mlcbakery/models.py) - SQLAlchemy ORM models for database entities - Database Layer (
mlcbakery/database.py) - Database connection and session management - Auth Layer (
mlcbakery/auth/) - Multiple authentication strategies (JWT, Admin Token, Passthrough) - Storage Layer (
mlcbakery/storage/) - GCP storage integration for artifacts - MCP Server (
mlcbakery/mcp/) - Model Context Protocol server implementation
Key architectural patterns:
- Async/await throughout for non-blocking I/O
- Dependency injection for database sessions and authentication
- Pydantic for data validation
- SQLAlchemy with async support for database operations
- OpenTelemetry instrumentation for observability
Development Commands
Setup and Installation
# Install uv (Python package manager)
curl -LsSf https://astral.sh/uv/install.sh | sh
# Set up environment variables
cp env.example .env
# Install Python 3.12 if needed
uv python install 3.12
# Install dependencies (use uv sync for more reliable installation)
uv sync --python 3.12
Running the Application
Local Development:
# Run the FastAPI server with auto-reload
uv run uvicorn mlcbakery.main:app --reload --host 0.0.0.0 --port 8000
Docker Development:
# Start all services (postgres, typesense, api, mcp, caddy reverse proxy)
docker compose up -d
# Create database and run migrations
docker compose exec db psql -U postgres -c "create DATABASE mlcbakery;"
docker compose exec api alembic upgrade head
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 169 lines · 1,339 tokens per session scan C eb70c567b438
mlcbakery CLAUDE.md is an instructions file published in the GitHub repository jettyio/mlcbakery (7 stars, last pushed 5mo ago), licensed MIT. It adds 1,339 tokens to every session, about $0.0067 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
skills AGENTS.md
Instructions for qdrant/skills, covering qdrant skills, project overview, project structure, navigating skills locally and conventions.
seekstone CLAUDE.md
Instructions for shaqmughal/seekstone, covering claude.md, what this repo is, commands, the harness itself (run after npm install) and architecture.
marklogic-mcp CLAUDE.md
Instructions for tternquist/marklogic-mcp, covering marklogic mcp — agent working instructions, core principle: problem-first thinking, where guidance lives — skills first, agent skills (.claude/skills/ /skill.md) and marklogic://instructions resource (src/resources/index.ts).
plan-forge database.instructions.md
Database patterns for .NET — Dapper/EF Core, parameterized queries, migration strategy.
creditkarma-mcp CLAUDE.md
Claude Code instructions for chrischall/creditkarma-mcp, covering creditkarma-mcp, commands, tool naming, architecture and auth resolution (pattern a template).
copilot-instructions database.instructions.md
Instructions for ThiagoGuislotti/copilot-instructions, covering schema design, normalization, cartesian explosion, parameter sniffing and query performance.