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/kaju-open-source/ore/agents-mdgit clone --depth 1 https://github.com/Kaju-Open-Source/OREWhat 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 | $0.03562 | $0.03562 |
| Opus 5 | $0.01781 | $0.01781 |
| Sonnet 5 | $0.00712 | $0.00712 |
| Haiku 4.5 | $0.00356 | $0.00356 |
Grade A, and why
ORE AGENTS.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 — 344 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — project context for future assistants and contributors
Read this first when you (or an AI assistant) come back to this repo. It captures the current state and every gotcha so you don't have to rediscover them.
1. What this project is
ORE (Kaju Open Source) — a self-hostable AI-powered knowledge repository for academic documents. Three independent teams build it:
| Team | Folder | What they ship |
|---|---|---|
| 🤖 AI | ai-rag/ |
Python package: PDFs → embeddings |
| 🧠 Backend | backend/ |
FastAPI HTTP API wrapping the AI package |
| 🎨 Frontend | frontend/ |
Next.js 16 + React 19 + Tailwind v4 + TS |
Current state (early alpha, June 2026):
- ✅ AI team has shipped 3 modules:
ingestion,chunking,embedding - ✅ 10 offline pytest tests passing; 3 marked
@pytest.mark.network(need Ollama) - 🟡 Backend has a scaffold (FastAPI app + 1 smoke test) — not connected to AI yet
- 🟡 Frontend is the default
create-next-appboilerplate — not customized yet
2. Exact file layout (don't reorganize)
ore/ ← repo root
│
├── README.md 38 lines ← top-level orientation
├── CONTRIBUTING.md 81 lines ← team rules
├── COMMANDS.md ~186 lines ← every shell command + troubleshooting
├── AGENTS.md ← this file
├── LICENSE
├── .env.example ← ONE env var: ORERAG_EMBED_MODEL
├── .gitignore
│
├── docs/
│ ├── GETTING-STARTED.md 57 lines
│ └── architecture.png
│
├── ai-rag/ 🤖 AI team's home
│ ├── ingestion.py 📄 PDF → text (PyMuPDF)
│ ├── chunking.py ✂️ text → chunks (LangChain RecursiveCharacterTextSplitter)
│ ├── embedding.py 🧠 text → vectors (Ollama)
│ ├── pyproject.toml name = "orerag", pythonpath = ["."], ruff config
│ ├── README.md 50 lines (TL;DR + per-team specifics)
│ ├── CONTRIBUTING.md 3 lines (redirect to root)
│ ├── .gitignore
│ ├── tests/ ← ALL tests live here
│ │ ├── conftest.py
│ │ ├── data/sample.pdf ← offline test fixture
│ │ ├── test_ingestion.py
│ │ ├── test_chunking.py
│ │ └── test_embedding.py (3 tests marked @pytest.mark.network)
│ └── documents/ (empty placeholder for user uploads)
│
├── backend/ 🧠 Backend team's home
│ ├── app/main.py ← FastAPI app (scaffold: /, /health)
│ ├── tests/test_smoke.py ← one passing test
│ ├── pyproject.toml name = "orerag-backend"
│ ├── README.md 18 lines
│ └── .gitignore
│
└── frontend/ 🎨 Frontend team's home
├── app/ (page.tsx, layout.tsx, globals.css, favicon.ico)
├── public/ (5 default Next.js SVGs)
├── package.json / package-lock.json / tsconfig.json / …
├── README.md 15 lines
└── .gitignore
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 · 344 lines · 3,562 tokens per session scan A 9e69dcab3f6b
ORE AGENTS.md is an instructions file published in the GitHub repository Kaju-Open-Source/ORE (2 stars, last pushed 3d ago), licensed MIT. It adds 3,562 tokens to every session, about $0.0178 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 2026-08-31.
Other instructions, from other repositories
book-to-skill AGENTS.md
AGENTS.md instructions for virgiliojr94/book-to-skill, covering agents.md, project intent, sources of truth, non-negotiable rules and execution loop.
book-to-skill CLAUDE.md
Claude Code instructions for virgiliojr94/book-to-skill, a project described as: Turn any technical book PDF into a Claude Code skill — ready to study, reference, and use while you work.
MindTrain AGENTS.md
Instructions for shigella520/MindTrain, covering repository instructions, mindtrain target architecture, training and content integrity, safety and integrity and fast release workflow.
SkillNet AGENTS.md
Instructions for ANFAIA/SkillNet, covering agents.md, project, stack, repo structure and current phase: v1 and v2 always available.
classroomio CLAUDE.md
Claude Code instructions for classroomio/classroomio, a project described as: The Open Source Education Platform for Companies. A Simple and Beautiful Alternative to Moodle LMS, EdX, Thinkific and Teachable.
classroomio AGENTS.md
AGENTS.md instructions for classroomio/classroomio, covering agent guidance, research requirements, out of scope paths, translation, formatting, and git workflow and verification before commit/pr.