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/schmitech/orbit/claude-mdgit clone --depth 1 https://github.com/schmitech/orbitWhat 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 | $0.01234 | $0.01234 |
| Opus 5 | $0.00617 | $0.00617 |
| Sonnet 5 | $0.00247 | $0.00247 |
| Haiku 4.5 | $0.00123 | $0.00123 |
Grade A, and why
orbit CLAUDE.md scanned grade A with 0 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 2d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 113 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
ORBIT (Open Retrieval-Based Inference Toolkit) is a self-hosted AI gateway for production RAG (Retrieval-Augmented Generation). It exposes an OpenAI-compatible API that routes requests to 37+ LLM providers, performs intent-based SQL/NoSQL/vector retrieval, and includes enterprise controls (RBAC, audit logs, rate limiting, circuit breakers).
Commands
Setup
./install/setup.sh --profile default # Basic install
./install/setup.sh --profile all # All dependencies
./install/setup.sh --download-gguf [model] # Download a GGUF model
Running the Server
python3 server/main.py # Direct launch
python3 server/main.py --config /path/to.yaml # Custom config
./bin/orbit.sh start # Via shell wrapper
docker compose up -d # From docker/ directory
Testing
# From repo root, using the venv python:
/Users/remsyschmilinsky/Downloads/orbit/venv/bin/python -m pytest server/tests/
# Run a single test file:
/Users/remsyschmilinsky/Downloads/orbit/venv/bin/python -m pytest server/tests/test_foo.py
# Run by marker:
/Users/remsyschmilinsky/Downloads/orbit/venv/bin/python -m pytest server/tests/ -m unit
Pytest markers: unit, integration, slow. Python imports resolve from server/ — tests must be run with server/ as the working directory or the venv python from the repo root.
Linting
ruff check . # Check all Python files
ruff check server/ # Check server code only
Config: ruff.toml — line length 120, target Python 3.11+.
Architecture
Entry Points
server/main.py— FastAPI app factoryserver/inference_server.py— Server initialization, config loading, service wiring, lifespan manager
Service Layer (server/services/)
Services are initialized once via a service factory and injected via FastAPI dependency injection. Key services:
ChatService— message routing, streamingRetrievalService— RAG and document retrievalAPIKeyService— authentication and quota managementGuardrailService— content moderationCacheService— retrieval result cachingChatHistoryService— conversation persistence
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.
- 2d ago First seen · 113 lines · 1,234 tokens per session scan A 265d77068e1f
orbit CLAUDE.md is an instructions file published in the GitHub repository schmitech/orbit (343 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,234 tokens to every session, about $0.0062 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
LobsterAI AGENTS.md
AGENTS.md instructions for netease-youdao/LobsterAI, covering agents.md, instruction scope, project snapshot, cowork vs openclaw and commands.
hivekeep CLAUDE.md
Instructions for MarlBurroW/hivekeep, covering hivekeep, documentation map, tech stack, key conventions and naming.
hope-agent AGENTS.md
AGENTS.md instructions for shiwenwen/hope-agent, covering hope agent, 安全红线, 提交前检查（强制）, 分支与发布 and 设置约定.
goclaw AGENTS.md
AGENTS.md instructions for nextlevelbuilder/goclaw, covering goclaw gateway, language, tech stack, project structure and key patterns.
goclaw CLAUDE.md
Claude Code instructions for nextlevelbuilder/goclaw, covering goclaw gateway, language, tech stack, project structure and key patterns.
opencode-chat-bridge AGENTS.md
Instructions for ominiverdi/opencode-chat-bridge: You are a helpful assistant in a chat interface.