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 whenpoem/aiscientist --skill prove-sopgit clone --depth 1 https://github.com/whenpoem/aiscientistWrote 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/whenpoem/aiscientist/prove-sop)<a href="https://agentmods.dev/skills/whenpoem/aiscientist/prove-sop"><img src="https://agentmods.dev/badge/skills/whenpoem/aiscientist/prove-sop/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/whenpoem/aiscientist/prove-sop"><img src="https://agentmods.dev/badge/skills/whenpoem/aiscientist/prove-sop.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 | $0.00064 | $0.00453 |
| Opus 5 | $0.00032 | $0.00227 |
| Sonnet 5 | $0.00013 | $0.00091 |
| Haiku 4.5 | $0.00006 | $0.00045 |
Grade A, and why
prove-sop 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.
What it actually says
Prove SOP
Treat this skill as a router. Read references/proof-loop.md before starting a new proof or resuming an incomplete proof loop.
Route the request
- Identify the proposition, assumptions, requested rigor, and desired output.
- Look up related proof failures with
mcp__memory__match_signatures(..., domain="proof")before drafting. - Register the proposition if it is not already in the shared graph.
- Follow the proof loop reference through skeleton selection, draft diagnosis, correction, and review.
- Add an empirical companion only when a constant or premise needs data.
- Use Lean only as optional reinsurance for a small, closed, formalization-ready claim; a missing Lean tool or failed attempt does not invalidate a sound natural-language proof.
Required boundaries
- Compare only candidates of the same graph kind.
- Treat BT intervals as uncalibrated approximate posterior summaries. Never call their separation a significance test or use it as the sole stopping rule.
diagnose_snippetis read-only; persist each judgment explicitly withregister_diagnosis.- Corrected drafts need a new segmentation and diagnostic manifest. Preserve old manifests as audit history.
- Check wall-clock budget before a Lean attempt expected to take at least five minutes, and record actual use afterward.
- If the optional
proverorbudgeteragent is unavailable, perform the same checks inline and report that the specialized agent surface was absent.
Completion
Finish only when the proposition and final draft are identifiable, the latest diagnostic manifest is empty or every remaining limitation is explicit, the manuscript states Lean verification status honestly, and any empirical companion has passed its own preregistration and provenance gates.
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 · 44 lines · 64 tokens per session scan A 06acb1cbf54d
prove-sop is a skill published in the GitHub repository whenpoem/aiscientist (8 stars, last pushed 1mo ago), licensed MIT. It adds 64 tokens to every session and 453 once invoked, about $0.0003 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
anti-autoresearch
End-to-end substantive-integrity forensic sweep of a research paper (especially autoresearch / AI-Scientist-style output). Orchestrates the whole pipeline: ingest (arxiv-id | pdf | dir → working dir + pdftotext for L0) → /evidence-ledger (artifact manifest + observability level L0/L1/L2 + span-anchored claims.json) →…
baseline-comparison-audit
Audit whether a paper's baseline comparisons are COMPLETE, FAIR, and SIGNIFICANT: a required recent SOTA baseline is missing while 'best/SOTA' is claimed (HP-MISSING-BASELINE); a baseline is undertuned / given less compute-tuning-data, run at a mismatched config, or the equal-budget ablation-as-baseline is absent…
eval-design-forensics
Audit whether a paper's EVALUATION DESIGN actually measures what it claims and whether its reporting is complete — the validity layer family D (experiment-forensics) cannot reach. Three patterns: train/test leakage means the reported score may not measure generalization (HP-EVAL-LEAKAGE — adopts the Kapoor & Narayanan…
experiment-forensics
Audit experiment integrity against the evidence ledger. At L2 (repo + result files present) a fresh cross-model reviewer reads the eval code line-by-line for fake/derived ground truth, score self-normalization, phantom results (a paper number with no backing file/key), dead/uncalled metric code, verified-scope…
proof-derivation-forensics
Family-G proof & derivation integrity forensics: does a THIRD PARTY's written proof/derivation actually establish its theorem, or does it skip an obligation, assume its own conclusion, take an invalid step, drift a symbol's meaning, or smuggle an unstated assumption? Decides from the WRITTEN proof/derivation …
omnisci
Run OmniScientist end to end in the OmniScientist CLI using DeepSeek V4 Flash. Turn raw research data (images, signals, audio, video, 3-D, tables, or graphs) and an open direction into perceived evidence, a falsifiable hypothesis, recorded analysis, real citations, a gated candidate paper, PDF, and Overleaf bundle.…