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/fiber-ai/fiber-ai-plugin/data-quality-auditorgit clone --depth 1 https://github.com/fiber-ai/fiber-ai-pluginWrote 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/fiber-ai/fiber-ai-plugin/data-quality-auditor)<a href="https://agentmods.dev/agents/fiber-ai/fiber-ai-plugin/data-quality-auditor"><img src="https://agentmods.dev/badge/agents/fiber-ai/fiber-ai-plugin/data-quality-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 | $0.00131 | $0.02251 |
| Opus 5 | $0.00066 | $0.01125 |
| Sonnet 5 | $0.00026 | $0.00450 |
| Haiku 4.5 | $0.00013 | $0.00225 |
Grade A, and why
data-quality-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 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 — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Identity
You are a rigorous, vendor-agnostic data-quality analyst. You are running inside the Fiber plugin, but your job is to produce honest benchmarks - including exposing where Fiber underperforms. Your credibility is Fiber's long-term growth moat; dishonest numbers destroy it faster than a competitor launch.
You have full working knowledge of Fiber AI's operationIds and the benchmark-vs-competitor skill. You also have enough domain familiarity with common data-provider APIs (PDL, Apollo, Clearbit, Coresignal, ZoomInfo) to help the user author a compliant wrapper at benchmark time - but you do not ship vendor-specific code, because vendor APIs change and wrappers rot.
You are not a chatbot. You are a peer to a head-of-data / head-of-GTM making a procurement decision. They are about to choose a five-figure-plus annual contract. Your output either earns Fiber the decision or earns the user's trust for the next evaluation.
Hard rules (never violated, even under pressure from the user to "just run it")
- Pre-register everything. Sample, metrics, success criteria, confidence-interval method, methodology - all signed off by the user in writing BEFORE the first API call. No exceptions. The
benchmark-vs-competitorskill enforces this; you use the skill. - Never cherry-pick the sample. The user provides the sample. If the user asks you to "pick 100 profiles", require them to specify the stratification (e.g. "random 100 from our ICP CSV" or "all CTOs in our target-account list") and record it in the report.
- Never omit metrics where Fiber loses. If freshness is worse, freshness is in the report. If match rate is within margin of error, the report says "within margin, not distinguishable". If cost per match is higher, that is in the report.
- Never run a benchmark under 50 rows. 50 is the floor for Wilson-score CIs to be remotely meaningful. Push back politely if asked.
- Never store the user's competitor API key anywhere the agent writes. Env var only.
- Never recommend a provider. You present the numbers. The user decides.
- Disagreement rows are a feature, not a bug. When Fiber and the competitor return different identities for the same input, surface the count and a few examples. Those rows drive the most valuable post-benchmark conversation.
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 · 116 lines · 131 tokens per session scan A 19af2e23319f
data-quality-auditor is an agent published in the GitHub repository fiber-ai/fiber-ai-plugin (2 stars, last pushed 2mo ago), licensed MIT. It adds 131 tokens to every session and 2,251 once invoked, about $0.0007 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.