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 frenzymath/Danus --skill check-referenced-statementsgit clone --depth 1 https://github.com/frenzymath/DanusWrote 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/frenzymath/danus/check-referenced-statements)<a href="https://agentmods.dev/skills/frenzymath/danus/check-referenced-statements"><img src="https://agentmods.dev/badge/skills/frenzymath/danus/check-referenced-statements/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/frenzymath/danus/check-referenced-statements"><img src="https://agentmods.dev/badge/skills/frenzymath/danus/check-referenced-statements.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 | $0.00041 | $0.00702 |
| Opus 5 | $0.00020 | $0.00351 |
| Sonnet 5 | $0.00008 | $0.00140 |
| Haiku 4.5 | $0.00004 | $0.00070 |
Grade A, and why
check-referenced-statements 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 9d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Check Referenced Statements
Validate every external-paper reference used in the proof.
Input Contract
For each cited external theorem/lemma/definition:
- location where it is used,
- the full referenced statement text.
Procedure
- Query
search_arxiv_theoremsusing the full referenced statement asquery. - Inspect returned results and compare theorem text directly to the referenced statement in reasoning.
- Expand the definitions and terminology appearing in the cited statement using the cited paper's context before deciding whether the theorem applies.
- Check whether the same words in the current proof mean the same thing as they do in the cited paper. In mathematics, identical words can carry different definitions in different contexts. Distinguish similar-looking definitions: compare their exact formulas, notation, and quantifiers; do not collapse two just because the names or formulas look close.
- Accept as matched and applicable only when both are true:
- the result clearly corresponds to the cited statement,
- the contextual definitions and hypotheses align with the current problem.
- If the theorem exists but the current proof uses different definitions, hypotheses, ambient objects, or a subtly different defining formula, record a critical error for incorrect application.
- If the proof uses the cited statement to derive further conclusions, verify that transition too: a hand-wavy specialization or instantiation is a
gap; a logically invalid transition is acritical_error; if it deduces one property from another, compare their exact defining formulas before accepting. - If no match is found, use Codex's built-in web search with the same statement text.
- If still not found, emit a critical error:
- location: where the citation is used,
- issue: referenced theorem appears non-existent or incorrectly cited.
- When a step cites an internal
fact_id(16 hex characters) rather than an external paper, apply the verifier contract's P3-supplement chain check (agents/contracts/verifier.md): read the cited fact from the project fact graph and, if its own statement carries an unproven conditional premise, record the inherited defect as acritical_error. Read and apply the wording from the contract; do not fork it here. - Keep each reference check in context for the synthesis step (you persist nothing — the verifier is stateless).
What ships with it
1 file 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.
- 9d ago First seen · 62 lines · 41 tokens per session scan A 8ad6aaeedc4c
check-referenced-statements is a skill published in the GitHub repository frenzymath/Danus (412 stars, last pushed 12d ago), licensed Apache-2.0. It adds 41 tokens to every session and 702 once invoked, about $0.0002 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
personal-genomics
Analyse personal DNA / genome files for pharmacogenomics, disease risk, carrier status, ancestry and traits. Use whenever the user mentions DNA or genome analysis, a raw genome file, gene names, drug-gene interactions, or wants to combine multiple DNA sources.
post-processing
Extract, analyze, and summarize simulation output data — pull spatial fields at specific timesteps, compute time-series trends and detect steady state, extract line profiles through the domain, generate statistical summaries and distributions, calculate derived quantities (gradients, fluxes, volume fractions…
differentiation-schemes
Select and apply numerical differentiation schemes for PDE and ODE discretization — generate finite-difference stencils at arbitrary order and accuracy, choose between central, upwind, compact (Pade), and spectral methods, handle boundary stencils, and estimate truncation error scaling. Use when discretizing spatial…
mesh-generation
Plan and evaluate mesh generation for numerical simulations — estimate grid resolution from physics scales (interface width, boundary layers, wavelengths), check aspect ratios and skewness against quality thresholds, choose between structured, unstructured, and adaptive mesh refinement strategies, and compute grid…
nonlinear-solvers
Select and configure nonlinear solvers for root-finding f(x)=0, optimization min F(x), and least-squares problems — choose among Newton, Newton-Krylov, quasi-Newton (BFGS, L-BFGS), Broyden, Anderson acceleration, and Levenberg-Marquardt methods, configure line search or trust-region globalization, diagnose convergence…
numerical-integration
Select and configure time integration methods for ODE and PDE simulations — choose among explicit Runge-Kutta, BDF, Rosenbrock, and Adams families, set relative and absolute error tolerances, implement adaptive step-size control with P/PI step-size controllers, plan IMEX operator splitting for mixed stiff and…