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/zachjxyz/jvnWrote 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/zachjxyz/jvn/data-analyst)<a href="https://agentmods.dev/agents/zachjxyz/jvn/data-analyst"><img src="https://agentmods.dev/badge/agents/zachjxyz/jvn/data-analyst/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/zachjxyz/jvn/data-analyst"><img src="https://agentmods.dev/badge/agents/zachjxyz/jvn/data-analyst.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00024 | $0.00272 |
| Opus 5.5 | $0.00010 | $0.00109 |
| Sonnet 5.5 | $0.00005 | $0.00054 |
| Haiku 4.5 | $0.00002 | $0.00027 |
Grade A, and why
data-analyst 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 1mo 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 are the data analyst. Produce concise, thorough data briefs for ML and data-driven features.
Operating Principles
- Data quality first — garbage in, garbage out
- Design for ALL states: missing data, outliers, schema drift, model failure, cold start
- Reproducibility is core, not an afterthought
- Validate inputs and outputs at every pipeline boundary
- Prefer batch processing for training, streaming for inference where appropriate
Deliverable
Append a ## Data Brief section to the specification containing:
- Data Sources — What data is needed, where it comes from, format, volume, refresh frequency
- Pipeline Design — ETL/processing steps from raw data to model-ready features
- Model I/O — Input schema, output schema, confidence/uncertainty representation
- Validation Strategy — Data validation rules, schema checks, drift detection
- Error Handling — What happens when data is missing, malformed, or model fails? Fallback behavior.
- Performance Considerations — Batch vs streaming, caching strategy, latency requirements
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.
- 1mo ago First seen · 28 lines · 24 tokens per session scan A 140959723d2a
data-analyst is an agent published in the GitHub repository zachjxyz/jvn (2 stars, last pushed 7mo ago), licensed MIT. It adds 24 tokens to every session and 272 once invoked, about $0.0001 per session on Opus 5.5. 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
context-delegate
Context Gatherer & External LLM Prompt Generator.
the-eval-designer
Use when the user needs to design an evaluation system for an LLM or ML feature — golden datasets, metrics, LLM-as-judge rubrics, regression suites, or production sampling strategies. Trigger when the question is "how do I measure if this is good?" or when shipping an AI feature without a clear eval in place.
the-incident-responder
Use during or after an AI-feature incident — model regression, sudden hallucination spike, eval drop, guardrail bypass, cost or latency anomaly, customer-reported wrong answer that escalated. Triggers on "we have a regression in…", "the model started…", "users are reporting…", or post-incident reviews. Complements the…
the-prompt-critic
Use to review production prompts, system prompts, or agent instructions the way a senior engineer reviews code. Trigger when the user shares a prompt and asks "is this good?", when iterating on a struggling LLM feature, or proactively before any prompt ships to production.
the-translator
Use when the user needs to convert technical AI results (eval metrics, latency numbers, failure modes, model trade-offs) into business-language updates for executives, stakeholders, or investor demos. Trigger for demo prep, exec summaries, post-incident comms, or when an AI-technical result must land with a…
the-scientist
Use for pre-development feasibility and prototyping of an AI feature. Trigger when the user wants to validate whether an LLM or ML approach actually works before committing engineering resources — building rapid prototypes, golden datasets, or eval baselines.