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 agentmods add commands/cbirkbeck/mathlib-quality/unformalisegit clone --depth 1 https://github.com/CBirkbeck/mathlib-qualityWrote 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/commands/cbirkbeck/mathlib-quality/unformalise)<a href="https://agentmods.dev/commands/cbirkbeck/mathlib-quality/unformalise"><img src="https://agentmods.dev/badge/commands/cbirkbeck/mathlib-quality/unformalise.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 | $0.00155 | $0.04943 |
| Opus 5 | $0.00077 | $0.02472 |
| Sonnet 5 | $0.00031 | $0.00989 |
| Haiku 4.5 | $0.00015 | $0.00494 |
Grade A, and why
unformalise 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 4d 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 — 487 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/unformalise — Turn Lean into mathematics
Take a Lean declaration and render its mathematical statement plus a paragraph-level proof sketch in standard mathematical English. Default output is Unicode in the terminal — readable directly in chat, without a LaTeX renderer.
This is the conversational front-door for the same unformalisation logic that
/blueprint Phase 4 uses to author whole-project blueprints. Use /unformalise when
you want to see the math first; use /blueprint when you want to commit the math
to the project artifact.
Usage
/unformalise <Foo.bar> # single declaration, Unicode in terminal (default)
/unformalise <Foo.bar> --closure # decl + dependency closure (every {uses} target)
/unformalise <file.lean> # every public declaration in the file
/unformalise <Foo.bar> --verso # print Verso directives to stdout (no blueprint write)
/unformalise <Foo.bar> --md # print GitHub-flavoured Markdown to stdout
/unformalise <Foo.bar> --blueprint # append to <Project>/Chapters/ as Verso (same as /blueprint --decl)
/unformalise <Foo.bar> --statement-only # skip the proof sketch
After the terminal Unicode render, the command asks what you want next (blueprint /
Verso / Markdown / nothing). The --verso / --md / --blueprint flags skip that
prompt — useful when wiring this into a pipeline or another command.
Prerequisites
- Project is a Lean 4 project that builds clean (
lake buildexits 0). The skill useslean_local_search/Grepto locate declarations; a broken build leaves the LSP unable to resolve names. - For
--blueprintoutput: the verso-blueprint scaffold exists at the project root (theproject_template/layout fromleanprover/verso-blueprinthas been copied in —<Project>/Chapters/,<Project>/Blueprint.lean,<Project>Main.lean,scripts/ci-pages.sh). See/blueprintPhase 0 for bootstrap instructions. - For best quality:
.mathlib-quality/references/directory of source-paper notes and/or a.mathlib-quality/blueprint/prose_context.mdfrom a prior/blueprintrun. The skill works without them but leans more heavily on Lean docstrings.
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.
- 4d ago First seen · 487 lines · 155 tokens per session scan A 3ac57b9bbf8f
unformalise is a command published in the GitHub repository CBirkbeck/mathlib-quality (32 stars, last pushed 16d ago), licensed MIT. It adds 155 tokens to every session and 4,943 once invoked, about $0.0008 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 commands, from other repositories
master_analysis
Run comprehensive 5-phase analysis across labs, genetics, and protocols.
render-figures
Compile all .tex and .typ figure files in a directory.
replication-package
Scaffold or audit a social-science replication package at a target directory, and audit the manuscript and its archived research objects against FAIR principles.
diff
Quantitative volume comparison between a CadQuery model and a reference STEP file.
arg-diagram
ARG academic-paper diagram mode — standalone structural & conceptual diagram generation.
simulation-calibrator
Test and refine simulation accuracy with validation loops, bias detection, and continuous improvement frameworks.