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/quattro/jax-numerics-agent/agents-mdgit clone --depth 1 https://github.com/quattro/jax-numerics-agentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/quattro/jax-numerics-agent/agents-md)<a href="https://agentmods.dev/instructions/quattro/jax-numerics-agent/agents-md"><img src="https://agentmods.dev/badge/instructions/quattro/jax-numerics-agent/agents-md.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00262 | $0.00262 |
| Opus 5 | $0.00131 | $0.00131 |
| Sonnet 5 | $0.00052 | $0.00052 |
| Haiku 4.5 | $0.00026 | $0.00026 |
Grade A, and why
jax-numerics-agent 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 5d 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.
What it actually says
AGENTS.md
Bootstrap contract for coding agents in this repository.
Scope
This repo is a guidance package (skills, checklists, snippets), not a runtime library.
Source of truth
Canonical rules live in:
skills/jax_equinox_best_practices/SKILL.mdskills/project_engineering/SKILL.md
If this file conflicts with either skill file, the skill files take precedence.
Agent workflow
- Load the relevant skill file(s) first.
- For numerics deep-dives, load topic docs from
skills/jax_equinox_best_practices/references/as directed by the numericsSKILL.md. - Follow companion checklist/snippet references declared in each skill.
- Keep
checklists/*,snippets/*, andreferences/*links in skill docs valid. - Run
./scripts/validate_skill_links.shafter editing skills/checklists/snippets.
Global install note
When installed via scripts/install_skills_with_assets.sh, assets live under:
~/.codex/skills/assets/checklists/~/.codex/skills/assets/snippets/
Maintenance
Keep this file minimal. Do not duplicate rule catalogs here; update the skill files.
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.
- 5d ago First seen · 34 lines · 262 tokens per session scan A f552829ff6a1
jax-numerics-agent AGENTS.md is an instructions file published in the GitHub repository quattro/jax-numerics-agent (9 stars, last pushed 22d ago), licensed MIT. It adds 262 tokens to every session, about $0.0013 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
py3plex copilot-instructions.md
Copilot instructions for SkBlaz/py3plex, covering github copilot instructions for py3plex, project overview, repository structure, key code patterns and network creation.
viral-ngs copilot-instructions.md
Copilot instructions for broadinstitute/viral-ngs, covering copilot instructions, quick reference, running tests and key files.
FaST-LMM AGENTS.md
Instructions for fastlmm/FaST-LMM, covering coding notes for agents, general policies, cross-repository release coordination, numerical correctness and error handling.
StatsPAI CLAUDE.md
Claude Code instructions for brycewang-stanford/StatsPAI, covering claude.md — statspai, 1. 项目定位, 2. 仓库结构, 3. 设计原则 and 4. 代码规范.
panoptes-utils AGENTS.md
AGENTS.md instructions for panoptes/panoptes-utils, covering ai agent guidelines for panoptes utilities, project overview, essential reading, project structure and development workflow.
AWETrim AGENTS.md
Instructions for awegroup/AWETrim, covering awetrim — agent context, what this repo is, module status, physics references and symbol ↔ code name table.