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 skills add tetracoralla/math-anchor --skill calculategit clone --depth 1 https://github.com/tetracoralla/math-anchorWrote 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/skills/tetracoralla/math-anchor/calculate)<a href="https://agentmods.dev/skills/tetracoralla/math-anchor/calculate"><img src="https://agentmods.dev/badge/skills/tetracoralla/math-anchor/calculate/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tetracoralla/math-anchor/calculate"><img src="https://agentmods.dev/badge/skills/tetracoralla/math-anchor/calculate.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00083 | $0.01123 |
| Opus 5 | $0.00042 | $0.00562 |
| Sonnet 5 | $0.00017 | $0.00225 |
| Haiku 4.5 | $0.00008 | $0.00112 |
Grade A, and why
calculate 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 6d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Calculate
Decide and route
- Do not load it for trivial, low-risk arithmetic the model can immediately verify.
- Treat fixed-width wrapping/saturating arithmetic as mandatory; do not classify them as trivial arithmetic.
- Always use it for fixed-width or IEEE-754 facts, exact/high-precision or symbolic work, matrices, units, statistics, probability, finance, verification, repeated calculations, or consequential results.
- Ask for a missing assumption when it changes the problem; a call must not hide ambiguity.
- The four tools are registered. Never call
list_mcp_resources, inspect source, or use a shell to find them. Usemath.runfor a known operation; otherwise search, then describe only if needed. - For an obligation DAG, failures-only checkpoint, or receipt replay, do not
search the MCP catalog. When this installed Skill contains the Host-generated
scripts/math-anchorlauncher, invoke it relative to this Skill directory withcheck-obligationsorreplay-obligations; retain receipts outside model context. Never substitute an ambient command or source checkout. If the launcher is absent, report that this carrier has no obligation checkpoint. - Every run uses the outer envelope
{operation, arguments}. Put all operation-specific fields insidearguments; never flatten them besideoperation.
Call these fully known routes directly; never search or describe these shapes first:
- Fixed-width arithmetic:
{"operation":"integer.machine_arithmetic","arguments":{"action":"add","left":"255","right":"2","bitWidth":8,"signedness":"unsigned","inputMode":"value","overflowBehavior":"wrapping"}} - Exact combinations/permutations:
{"operation":"combinatorics.count","arguments":{"action":"binomial","n":52,"k":5}} - Checkable polynomial identity:
{"operation":"certificate.polynomial_identity","arguments":{"left":"(x+1)^2","right":"x^2+2*x+1","variables":["x"]}}
Execute economically
- Use one
math.run,math.batchfor 2–32 independent ordered calculations, orfunction.samplefor one expression at many points. Batch is not a dependency graph. - For bulky output, select
resultModeand raisemaxOutputBytesonly when useful. - Control digits with
arguments.precision;precisionis not a top-levelmath.runfield. It counts significant digits. Exact fields stay exact. For decimal places, include integer digits plus at least two guard digits in the first call and round once for presentation. - Supply answer-changing variables, bounds, units, conventions, and tolerances.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago Changed · -7 lines · -26 tokens per session 5ce24ddbcf25
- 10d ago First seen · 93 lines · 109 tokens per session scan A bbda7437b567
calculate is a skill published in the GitHub repository tetracoralla/math-anchor (0 stars, last pushed 4d ago), licensed Apache-2.0. It adds 83 tokens to every session and 1,123 once invoked, about $0.0004 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 skills, from other repositories
journal-cover-letter-skill
Create, revise, benchmark, or check a cover letter for an academic journal submission from manuscript files. Use whenever a user wants a journal submission letter, asks to tailor a cover letter to a target journal, provides manuscript materials and asks for a letter, wants an existing academic cover letter improved…
journal-cover-letter-skill-trainer
Improve, extend, or benchmark the journal-cover-letter-skill from paired manuscript materials and expert-authored cover letters. Use whenever a user wants to train or iterate the Cover Letter Skill, compare an AI letter with an expert letter, strengthen the Research, Review, or Bibliometrics route, add a new writing…
julia-expert
Expert knowledge in Julia scientific computing, multiple dispatch, performance optimization, and numerical analysis. Use when the user mentions scientific computing, multiple dispatch, performance, numerical analysis, data science, or HPC, or when the task involves Type System, Array Programming, Installation and…
or-solver
Unified solver detection, installation, selection, and license configuration for OR optimization skills (LP, MIP, SOCP). One-stop solver management: detect installed packages, install missing solvers with proper priority, select the best available solver by problem type, and troubleshoot common errors. Covers 20…
cdopt-optimization
Use when Codex needs to solve, reproduce, test, compare, generate examples for, or diagnose CDOpt manifold optimization workflows, including CDOpt package smoke tests, official problem description cards, Stiefel dictionary learning, SciPy wrappers, PyTorch/JAX/NumPy CDOpt examples, dependency checks, tiny CPU…
linear-programming
Models and solves linear programs (LP) with multiple solvers (COPT, Gurobi, MOSEK, CPLEX, HiGHS (scipy/highspy), CLARABEL, PuLP/CBC, CVXOPT, OR-Tools/GLOP, GLPK, SoPlex, lpsolve, ECOS). Restates natural-language word problems into LP before building and solving. When no local solver is available, searches GitHub for…