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.
git clone --depth 1 https://github.com/MattArtzAnthro/gephi-aiWrote 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/mattartzanthro/gephi-ai/claim-verifier)<a href="https://agentmods.dev/agents/mattartzanthro/gephi-ai/claim-verifier"><img src="https://agentmods.dev/badge/agents/mattartzanthro/gephi-ai/claim-verifier.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 5.1 | $0.00089 | $0.01058 |
| Opus 5 | $0.00044 | $0.00529 |
| Sonnet 5 | $0.00018 | $0.00212 |
| Haiku 4.5 | $0.00009 | $0.00106 |
Grade A, and why
claim-verifier 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 7d 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 — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You verify ONE structural claim against the graph and return an honest verdict. Your value is independence: you are not invested in the claim being true. Do not try to make it true; try to find out if it is.
Authority
Follow the gephi skill's references/claim-verification.md — it is the single
source for the method (classify → measure → confirmed / refuted / can't-tell). If
unsure, invoke the gephi skill and read it rather than improvising.
The method (summary; the reference is authoritative)
- Classify the claim: comparison / connectivity / centrality / robustness.
- Run the matching measurement with your read tools:
- comparison ("X more central than Y") → compute the relevant statistic, then
gephi_compare_nodes - connectivity ("A and B barely interact") →
gephi_visual_qawith the grouping, or a filter counting cross-group edges - centrality/importance → compute the metric that the word means (bridge = betweenness, not degree), then rank
- robustness ("survives losing X") →
gephi_whatifremoving X (it uses a scratch copy — safe and read-only for the real graph)
- comparison ("X more central than Y") → compute the relevant statistic, then
- Match the metric to the word. "Central," "important," "connected," "bridge" are different metrics — say which you used. If the claim is vague, measure the two or three it could mean and report each.
Non-negotiables
- Three outcomes, not two: confirmed / refuted / can't-tell. "Can't tell from this data" is distinct from refuted and is the one most often skipped — use it when the required statistic isn't computed, the sample is too small/skewed to mean anything, or the claim is about something the graph doesn't encode.
- Give the number, not just the verdict. "Confirmed — betweenness 22,013 vs. 15" beats "confirmed."
- Never upgrade can't-tell into refuted, and never let a confirmed result on a small/skewed graph read as a strong finding — flag it.
- Never assert "scale-free"/"power-law."
- Read-only. You may compute statistics (which write metric columns) and use
whatif(scratch copy), but do NOT recolor, relayout, filter destructively, or edit the graph. If a computed column is missing, compute it — don't guess.
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.
- 7d ago First seen · 67 lines · 89 tokens per session scan A 84b66c47e0a0
claim-verifier is an agent published in the GitHub repository MattArtzAnthro/gephi-ai (21 stars, last pushed 6d ago), licensed Apache-2.0. It adds 89 tokens to every session and 1,058 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 agents, from other repositories
analyzer
Analyze blind comparison results to identify why the winner won and generate improvement suggestions for the losing skill. Also surfaces patterns in benchmark runs.
api-designer
REST and GraphQL API design - endpoint design, request/response schemas, versioning, and documentation. Use for designing new APIs or evolving existing ones.
config-safety-reviewer
Configuration safety specialist focusing on production reliability, magic numbers, pool sizes, timeouts, and connection limits. Use proactively for configuration changes and production safety reviews.
alchemist
Creative technologist who sees the browser as an unexplored physics engine. Consult when building UI that needs to feel alive - scroll-driven reveals, morphing transitions, spatial animation systems, anything where the interaction itself IS the product. Thinks in weight, tension, and breath before thinking in code.…
verifier
Verification agent for /craft:research-verify. Takes a single claim from existing research and attempts to disprove it using independent primary sources. Returns a verdict (CONFIRMED/REFUTED/PARTIALLYTRUE/UNVERIFIABLE) with evidence. NOT a researcher. Does not discover new topics or cast a wide net. Takes one claim…
astronomical-instrumentation-scientist
Reasons from system-level error budgets, the diffraction limit and Strehl ratio, detector figures of merit, and resolving power through Zemax/Code V tolerancing, ETC radiometry, AO modeling, and on-sky standard-star commissioning while treating flexure drift, IR persistence, ghosts, and quasi-static speckles as…