Borrowing it
Nothing to install: this file belongs to danielsimonjr/math-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/danielsimonjr/math-mcp/master/CLAUDE.mdgit clone --depth 1 https://github.com/danielsimonjr/math-mcpWrote 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/danielsimonjr/math-mcp/claude-md)<a href="https://agentmods.dev/instructions/danielsimonjr/math-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/danielsimonjr/math-mcp/claude-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 | $0.02521 | $0.02521 |
| Opus 5 | $0.01260 | $0.01260 |
| Sonnet 5 | $0.00504 | $0.00504 |
| Haiku 4.5 | $0.00252 | $0.00252 |
Grade A, and why
math-mcp 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 yesterday.
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 — 219 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.
v4 — MathTS engine (READ FIRST)
As of v4.0.0 the compute engine is MathTS (@danielsimonjr/mathts-*), not
mathjs. The 7 MCP tools keep identical I/O contracts; only the internals changed.
v4.1.x has a real solve: exact roots for polynomials of degree ≤ 3
(incl. complex) and a numeric real-root fallback for degree ≥ 4 / transcendental
equations. The root-finding lives in MathTS as a first-class function —
math.solve(equation, variable) in @danielsimonjr/mathts-functions — and
handleSolve only validates input, detects degenerate cases, and formats the
roots. As of 4.1.2 (MathTS functions ≥ 0.2.10) math.solve delegates its
degree-≤3 closed form to the engine's Algebra solver polynomialRoot.
Correction (2026-06-23): earlier notes here claimed
polynomialRootwas avoided due to a forked-typed-function"nested-dispatch" bug. That was a misdiagnosis. typed-function was correct. The real bug was MathTS'sadd/multiplydeclaring only anumber-variadic, soadd(number, Complex, Complex)(polynomialRoot'sadd(b, C, …)with a complex cube root) had no matching signature. Fixed in@danielsimonjr/[email protected]by making the variadics'any, any, ...any';polynomialRootnow works andsolveuses it. (A minor, non-blocking gap remains:cbrt(number, allRoots=true)is still unimplemented; theComplexallRoots path polynomialRoot needs works.)
- Engine:
src/math-engine.tsbuilds the instance via@danielsimonjr/mathts-compatcreate(all)(mathjs-compatible API). Every module imports its default export. mathjs and the hand-rolled acceleration stack (router/adapter/wasm-executor/gpu/workers, thewasm/AS project) were deleted — MathTS does its own internal tier dispatch. Entry point isdist/index.js(the oldindex-wasm.jsis gone). - Dependencies (hybrid): Bun is the install/script toolchain (
bun.lockauthoritative). Dev can use local links —bun add file:../mathts/compat file:../mathts/matrix— so changes to the local~/Github/mathtsmonorepo are picked up after abun installrefresh +bun run build. For a committed/published release, pin the published@danielsimonjr/mathts-*@^x.y.zversions (after the MathTS packages are published to npm) andbun install. - Known limitations:
- Large dense matrices are slower than the old WASM path: MathTS's JS matmul does ~21s for 800×800 (was 114s before a perf fix; Rust WASM is not yet built/packaged). Small matrices — the normal tool use — are fast. DoS for huge inputs is enforced by size limits, not timeouts (synchronous JS can't be interrupted).
- The old mathjs fork's custom astronomical/nautical/typography units
(
lightyear,parsec,AU,nauticalMile,fathom, …) are not in MathTS's unit set; the unit tool returnsUnit "X" not found.for them. - Speed shorthands
mph/kph/knotare not units — use the compound formsmi/handkm/hinstead. This is not a cutover regression: stock mathjs (verified against v15.0.0) also errors on these (Undefined symbol mph); theunit_conversiondescription used to give'mph'as an example, which never worked on either engine — corrected to'mi/h'on 2026-06-23.
- Build is
bun run build(tsc) only. There is nowasm/project and nobuild:wasm/build:all/generate:hashesscripts — they were removed with the accel stack in v4. (CI's obsoletecd wasmstep was likewise dropped 2026-06-30.) - Runtime: Node ships the MCP server (
node dist/index.js). Bun is the TypeScript-on-Bun toolchain only (install +bun runscripts) — same split as the sibling*-mcprepos.
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.
- yesterday Changed · +5 lines · +93 tokens per session 728bc52f19e1
- 7d ago First seen · 214 lines · 2,428 tokens per session scan A 45bb30019554
math-mcp CLAUDE.md is an instructions file published in the GitHub repository danielsimonjr/math-mcp (0 stars, last pushed today), licensed ISC. It adds 2,521 tokens to every session, about $0.0126 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 2026-09-01.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.