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/project-statusgit clone --depth 1 https://github.com/CBirkbeck/mathlib-qualityWhat 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 | $0.00080 | $0.06210 |
| Opus 5 | $0.00040 | $0.03105 |
| Sonnet 5 | $0.00016 | $0.01242 |
| Haiku 4.5 | $0.00008 | $0.00621 |
Grade A, and why
project-status 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 2d 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 — 540 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/project-status — What is the worker doing, and how does it fit?
A chat-only status report. The agent reads the project's .lean code (and
.mathlib-quality/plan.md and .mathlib-quality/tickets.md if they exist),
un-formalises into mathematical English, and answers four questions:
- What mathematical result is the worker currently on?
- What (if anything) is blocked, and what is missing?
- How does the current work connect to the project's overall goal?
- How far along is the whole project?
The audience is an expert mathematician glancing between work sessions. The report is mathematical reportage — not Lean tactics, not difficulty rhetoric.
This command is read-only on every file. It does not start servers, open browsers, or write sidecar JSONs.
Self-contained for a cold reader (binding principle)
The reader has not followed the formalisation closely and cannot open any file. Everything they need to understand the state of play must be in the report itself. Concretely:
-
Define what is non-standard. The first time a project-specific notion appears (
eisensteinSeries,cauchyPrincipalValue,fooBar, etc.), state its mathematical content in one phrase: "the Eisenstein series E_k(τ) = ∑_{(c,d) ≠ 0} (cτ + d)^(−k)", not "eisensteinSeries (the project's main object)". -
Translate structural moves. When the proof body uses Lean's structural tactics (
obtain,refine,constructor,have,use), do not say "the worker obtains a witness" or "splits into cases". Say what the witness is, what the cases are, what the parametric intermediate claim asserts. The reader doesn't know Lean tactics; they know mathematics. -
Spell out parametric / universally-quantified intermediate results. When the proof builds a fact of shape "for every ε > 0 there is N such that …", state the universal claim explicitly. Don't hide it behind a name like
h_cauchy. -
Name the witnesses. When the proof chooses a specific bound, constant, or function, say what it is. "The worker chose M = π²/6 as the upper bound, motivated by ∑ n⁻² = π²/6." Not "fixes a constant".
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.
- 2d ago First seen · 540 lines · 80 tokens per session scan A ea18f0c422a2
project-status is a command published in the GitHub repository CBirkbeck/mathlib-quality (32 stars, last pushed 13d ago), licensed MIT. It adds 80 tokens to every session and 6,210 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 2026-08-30.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.