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/morluto/jacobian/agents-mdgit clone --depth 1 https://github.com/morluto/jacobianWrote 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/morluto/jacobian/agents-md)<a href="https://agentmods.dev/instructions/morluto/jacobian/agents-md"><img src="https://agentmods.dev/badge/instructions/morluto/jacobian/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 | $0.03192 | $0.03192 |
| Opus 5 | $0.01596 | $0.01596 |
| Sonnet 5 | $0.00638 | $0.00638 |
| Haiku 4.5 | $0.00319 | $0.00319 |
Grade A, and why
jacobian 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 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.
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 — 282 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Jacobian agent guide
This file states the product choices and repository invariants that changes must preserve. Follow CONTRIBUTING.md when selecting validation, changing documentation, preparing commits or pull requests, or working on releases and evaluations. Follow the product model and architecture when changing product scope, ownership, or the execution path.
What we are building
Jacobian gives agents atomic, composable tools for higher mathematics: discovering, running, and combining typed computations to investigate conjectures, build examples, calculate invariants, and check bounded claims.
Jacobian's hypothesis is that mathematical reasoning benefits from an executable vocabulary of semantically scoped, bounded operations. Prefer reusable mathematical primitives over large solvers or workflows: Jacobian supplies the mathematical moves; the model decides how to compose them into larger solutions.
Atomicity is semantic: an operation establishes one stable, reusable mathematical postcondition. It need not have a small or simple implementation.
Established mathematical techniques are sources of executable vocabulary, not automatic operation IDs. Make a technique discoverable through relevant public operations, but admit it separately only when it establishes a distinct stable postcondition. Alternative algorithms for the same postcondition remain private kernel choices.
It exposes two MCP tools:
| Agent verb | MCP tool | Meaning |
|---|---|---|
| Search | math.find |
Find or inspect an operation. |
| Execute | math.run |
Run one operation and return its mathematical value. |
Jacobian supplies bounded typed operations and immutable discovery. Use “operation” or “math tool,” not “product” or “provider,” for built-ins. It is local-first: ordinary mathematical tool work should stay focused on mathematics; preserve an explicit transport/security boundary only when the task actually changes one.
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 · 282 lines · 3,192 tokens per session scan A 6ed76d3d0b1c
jacobian AGENTS.md is an instructions file published in the GitHub repository morluto/jacobian (92 stars, last pushed 3d ago), licensed MIT. It adds 3,192 tokens to every session, about $0.0160 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
OpenQuantum AGENTS.md
Instructions for xi-zhao/OpenQuantum, covering openquantum, 四个职责面, 扩展对象不可混用, 不变量 and 代码位置.
sagemath-mcp CLAUDE.md
Instructions for XBP-Europe/sagemath-mcp, covering claude.md, project, commands, linting and testing.
chemdraw-mcp CLAUDE.md
Instructions for jurimaxam-dotcom/chemdraw-mcp, covering chemdraw-mcp, build / run, auto-gate (the project's green/red), the five tool areas (where a new tool goes) and architecture.
cdc-health-mcp-server CLAUDE.md
Instructions for cyanheads/cdc-health-mcp-server, covering agent protocol, domain, api surface, socrata api endpoints and cdc wonder api.
aurora-agent AGENTS.md
AGENTS.md instructions for AURORA-NEURO/aurora-agent, covering aurora agent, what we can never compromise on, 1. honest labelling is the product, 2. invariants belong in the type system, not in comments and 3. determinism, byte for byte.
hea-bench AGENTS.md
AGENTS.md instructions for dfieser/hea-bench, covering agents.md, what this is, install and import, the one mental model you need and descriptors (pure functions, exact units).