Borrowing it
Nothing to install: this file belongs to morluto/jacobian. 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/morluto/jacobian/main/.agents/skills/audit-mathematical-vocabulary/SKILL.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/skills/morluto/jacobian/audit-mathematical-vocabulary)<a href="https://agentmods.dev/skills/morluto/jacobian/audit-mathematical-vocabulary"><img src="https://agentmods.dev/badge/skills/morluto/jacobian/audit-mathematical-vocabulary/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/morluto/jacobian/audit-mathematical-vocabulary"><img src="https://agentmods.dev/badge/skills/morluto/jacobian/audit-mathematical-vocabulary.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00028 | $0.01137 |
| Opus 5 | $0.00014 | $0.00568 |
| Sonnet 5 | $0.00006 | $0.00227 |
| Haiku 4.5 | $0.00003 | $0.00114 |
Grade A, and why
audit-mathematical-vocabulary 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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audit Mathematical Vocabulary
Find confirmed gaps in one mathematical slice without turning the audit into a backend-wrapper wishlist or a claim of complete coverage.
Use the relevant vocabulary model
and domain references to assess the slice. Consult the
admission contract
when proposing publication or changing an execution envelope. Use
audit-public-operation-contracts for a deep review of one operation and
recent-conjecture-evaluations for a held-out conjecture evaluation.
Audit one slice
Choose one bounded slice, such as polynomial elimination. Name its mathematical boundary and the current-main revision. Do not silently expand it into a whole domain or repository audit.
Snapshot the relevant catalog facts before researching solutions. Record the operation IDs, owner domains, canonical inputs and outputs, natural downstream consumers, defining or reconstruction invariants, admitted bounds, and private backends. Use the live catalog and current source as authority; search results alone are not a complete inventory.
Check the slice through four independent lenses:
- Source demand. Use primary sources or established exact tasks to identify a finite mathematical obligation. State the needed result without importing the source's proof strategy. Literature recall or one bespoke workflow is not enough evidence for a public operation.
- Composition. Test whether producer values can enter their natural consumers unchanged after serialization. Include the empty, zero, singular, or identity case most likely to lose parent, axes, multiplicity, witness, or ambient context.
- Contract. Check accepted input, work and intermediate growth, output bounds, exact-result reconstruction, source binding, and typed failure behavior. Separate the semantic mathematical domain from the current execution envelope, and identify the quantities that actually control work and result size. A timeout, unavailable backend, or incomplete search is not a mathematical conclusion.
- Backend feasibility. Check maintained exact backends for an appropriate kernel only after establishing mathematical demand. Compare compact exact representations and algorithm regimes before treating a small fixed cap as necessary. Backend availability is evidence of feasibility, not evidence that Jacobian needs another public operation.
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 · -13 lines · -45 tokens per session 57bda70a657d
- 12d ago First seen · 124 lines · 73 tokens per session scan A b7d2fa4d8727
audit-mathematical-vocabulary is a skill published in the GitHub repository morluto/jacobian (191 stars, last pushed yesterday), licensed MIT. It adds 28 tokens to every session and 1,137 once invoked, about $0.0001 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 skills, from other repositories
paper-search
Search, download, and read academic papers from 20+ sources (arXiv, PubMed, Semantic Scholar, CrossRef, etc). Use when the user asks to find papers, search for research, look up academic literature, download a paper PDF, or extract text from a paper.
onescience-research-workflow
A research-workflow planning skill for OneScience, a system that coordinates scientific computing tasks. It creates detailed plans for fields such as weather, biology, materials, and fluid simulations, including steps, dependencies, resources, checks, and fallback paths.
onescience-modelscope-publish
A tool for preparing a reproduced research model for ModelScope, a platform for sharing machine-learning models. It organizes files into the platform’s expected directory structure and creates upload guidance.
onescience-cli
A skill for running OneScience command-line commands on remote SSH or SLURM systems. SSH is a way to use another computer remotely, while SLURM schedules jobs on shared computing clusters.
onescience-data-analyzer
A data-analysis skill for OneScience, a research-focused AI platform. It accepts task details from the conversation or from handoff files, then carries out analysis, charts, and report generation.
onescience-infer
A workflow tool for running scientific machine-learning models. It helps prepare inputs, load models or saved checkpoints, run predictions, check the results, create visualizations, and compare them with a baseline.