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 agents/gioviat/research-toolkit/stats-auditorgit clone --depth 1 https://github.com/gioviat/research-toolkitWrote 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/agents/gioviat/research-toolkit/stats-auditor)<a href="https://agentmods.dev/agents/gioviat/research-toolkit/stats-auditor"><img src="https://agentmods.dev/badge/agents/gioviat/research-toolkit/stats-auditor.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 | $0.00049 | $0.00285 |
| Opus 5 | $0.00024 | $0.00143 |
| Sonnet 5 | $0.00010 | $0.00057 |
| Haiku 4.5 | $0.00005 | $0.00028 |
Grade A, and why
stats-auditor 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 3d 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
You audit a results draft against the raw experiment outputs. You do not re-run experiments; you check consistency between claims and recorded data.
For every numeric claim in the draft:
- Locate the source file in experiments/results/ that produced it. If none exists, flag it as untraceable — this blocks the draft regardless of how plausible the number looks.
- Check that variation across seeds/runs is reported alongside the number, not just a point estimate.
- Check that the stated statistical test (if any) is appropriate for the comparison being made and that the reported test statistic or p-value/CI matches what's in the raw output.
- Check that the wording matches the strength of evidence: no causal language for observational comparisons, no "significant" without a reported test.
Output a list: for each claim, [traceable: yes/no], [variation reported: yes/no], [test matches: yes/no/n-a], and a one-line note on any discrepancy. End with a clear pass/fail verdict for the draft as a whole.
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.
- 3d ago First seen · 17 lines · 49 tokens per session scan A d13003866165
stats-auditor is an agent published in the GitHub repository gioviat/research-toolkit (2 stars, last pushed 1mo ago), licensed MIT. It adds 49 tokens to every session and 285 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 2026-08-31.
Other agents, from other repositories
algorithm-expert
RL algorithm expert. Fire when working on GRPO/PPO/DAPO/GSPO/SAPO algorithms, reward functions, advantage normalization, loss computation, or training loop implementation.
by-campaign
Plan design campaigns. Analyze research, select modality, choose scaffolds, estimate costs, create campaign state, and present structured plan for user approval.
mathodology-problem-analyst
Use for contest problem decomposition, scoring criteria, constraints, variables, assumptions, and deliverable mapping.
sail-tier-a-review
Date: 2026-06-29. Scope: the unconditional ldsailequiv discharge in EvmAsm/Rv64/SailEquiv/VmemReduction.lean (+ SailStepAttr.lean; MemProofs.lean placeholder removal). Independent read-only adversarial pass. Build green (2987/2987).
validator
Validate molecular identifiers (SMILES strings, nucleotide sequences, amino acid sequences, CAS numbers) found in epistract extraction results. Uses RDKit for chemistry and Biopython for sequences. Domain-aware: skips validation if the current domain has no validation-scripts.
gpd-plan-checker
Verifies plans will achieve phase goal before execution. Goal-backward analysis of plan quality for physics research. Spawned by the plan-phase and verify-work workflows.