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/modernrelay/omnigraph/agents-mdgit clone --depth 1 https://github.com/ModernRelay/omnigraphWhat 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 | $0.01609 | $0.01609 |
| Opus 5 | $0.00805 | $0.00805 |
| Sonnet 5 | $0.00322 | $0.00322 |
| Haiku 4.5 | $0.00161 | $0.00161 |
Grade A, and why
omnigraph AGENTS.md scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`curl -sL <url> | pandoc -f html -t markdown`. How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OmniGraph agent guide
This file is the always-on map for coding agents. Keep it short: current detail
belongs in docs/, exact behavior in code and tests, decisions in RFCs, and
open work in issues.
Required reading
Before every task and every change:
- Read architectural invariants. Apply the hard invariants and deny-list to implementation and review work.
- Consult the domain map in Lance alignment. Fetch and
read the complete content of every page in the matching domain plus every
page that is even slightly relevant. The index and summaries are not
substitutes for the upstream pages. A reliable local form is
curl -sL <url> | pandoc -f html -t markdown. - Read testing. Find the existing owner and run a clean focused baseline before changing tests. Extend an existing assertion, fixture, or parameterization before creating a parallel setup.
Tools that support @ imports include these automatically:
@docs/dev/invariants.md @docs/dev/lance.md @docs/dev/testing.md
CLAUDE.md is a symlink to this file. Edit AGENTS.md only.
Repository snapshot
- Version surveyed: 0.10.0
- Rust stable, edition 2024; toolchain pinned in
rust-toolchain.toml - Storage substrate: Lance 10.0.0
- Workspace: compiler, storage, engine (
omnigraph-enginepackage), policy, API types, cluster, CLI, server, Azure admission wrapper, benchmark harness, andomnigraph-dst(deterministic simulation testing; needs--cfg tokio_unstable, set by its crate-local.cargo/config.tomlwhen cargo runs from the crate dir — compiles empty without it) - License: MIT
OmniGraph is a typed property-graph engine coordinating many versioned Lance
datasets. One graph commit publishes all participating table versions and graph
lineage together. It provides .pg schemas, .gq queries, graph branches,
three-way merge, vector/full-text search, Cedar policy, a CLI, and a cluster-only
HTTP server.
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 · 163 lines · 1,609 tokens per session scan A f23abee16e5f
omnigraph AGENTS.md is an instructions file published in the GitHub repository ModernRelay/omnigraph (1,085 stars, last pushed 2d ago), licensed MIT. It adds 1,609 tokens to every session, about $0.0080 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
m_flow AGENTS.md
AGENTS.md instructions for FlowElement-xinliuyuansu/m_flow, covering m-flow — developer & agent reference, 1. repository map, extension points, 2. local development and python backend (requires python 3.10 – 3.13).
samyama-graph CLAUDE.md
Instructions for samyama-ai/samyama-graph, covering claude.md, project overview, build & development commands, build and run tests (1814 unit tests).
LeanKG CLAUDE.md
Instructions for FreePeak/LeanKG, covering leankg - ai agent context, project overview, quick start, index a codebase and calculate impact radius.
LeanKG AGENTS.md
Instructions for FreePeak/LeanKG, covering leankg — agent context, build & test, cli quick reference, mandatory: docker mcp project paths and tool discovery prefer-order.
LeanKG GEMINI.md
Instructions for FreePeak/LeanKG, covering leankg - ai agent context, project overview, quick start, index a codebase and calculate impact radius.
kglite CLAUDE.md
Instructions for kkollsga/kglite, covering kglite — claude code conventions, build & test, architecture, the boundary principle (wrappers vs core) — summary and in-memory is the core product.