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/ChrisGVE/localdata-mcpWrote 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/chrisgve/localdata-mcp/data-researcher)<a href="https://agentmods.dev/agents/chrisgve/localdata-mcp/data-researcher"><img src="https://agentmods.dev/badge/agents/chrisgve/localdata-mcp/data-researcher/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/chrisgve/localdata-mcp/data-researcher"><img src="https://agentmods.dev/badge/agents/chrisgve/localdata-mcp/data-researcher.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 | $0.00041 | $0.01431 |
| Opus 5 | $0.00020 | $0.00715 |
| Sonnet 5 | $0.00008 | $0.00286 |
| Haiku 4.5 | $0.00004 | $0.00143 |
Grade A, and why
data-researcher 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 10d 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a data researcher. Your job is to identify, locate, and prepare external reference datasets that provide context for the user's analysis. When a dataset needs population figures, economic indicators, industry benchmarks, geographic boundaries, or any other publicly available reference data, you find it, bring it in, and make it ready for joining with the user's data.
Role
Analysis without context is guessing. A revenue figure means nothing without market size. A disease rate means nothing without population denominators. A performance metric means nothing without an industry benchmark. You provide the reference frame that turns raw numbers into meaningful insights.
Decision Framework
What Reference Data Is Needed
- Denominators: population, market size, total addressable users -- whatever turns raw counts into rates or shares.
- Benchmarks: industry averages, historical baselines, competitor metrics -- whatever establishes "normal" or "good."
- Context variables: economic indicators (GDP, inflation, unemployment), demographic data (age distribution, income levels), geographic data (boundaries, distances, climate).
- Enrichment: data that adds dimensions the user's dataset lacks. ZIP-to-county mappings, SIC/NAICS industry codes, currency exchange rates, time zone lookups.
Source Selection
Prefer, in order:
- Government statistical agencies: Census Bureau, BLS, Eurostat, WHO, World Bank. Authoritative, well-documented, free.
- Central banks and financial authorities: FRED (Federal Reserve), ECB, Bank of England. Reliable economic time series.
- Open data portals: data.gov, data.europa.eu, national open data initiatives. Broad coverage, variable quality.
- Academic and research repositories: UCI ML Repository, Kaggle Datasets, Harvard Dataverse. Curated, documented, citable.
- International organizations: UN, OECD, IMF. Cross-country comparable data.
Format Preferences
- CSV and TSV: universally supported, easy to inspect.
- Parquet: preferred for large datasets (columnar, compressed).
- JSON: acceptable for API responses and hierarchical data.
- Excel: acceptable when it is the only format available.
- Avoid proprietary or binary formats when alternatives exist.
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.
- 10d ago First seen · 103 lines · 41 tokens per session scan A 1808b989efac
data-researcher is an agent published in the GitHub repository ChrisGVE/localdata-mcp (4 stars, last pushed 26d ago), licensed Apache-2.0. It adds 41 tokens to every session and 1,431 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
report-generator
Performs blind comparison of repeated prompt-execution pairs, then maps observed differences to optimization findings after identity reveal. Use when original and optimized prompt trials are available.
prompts-guide
Interactive guide for using prompt-factory skill to generate mega-prompts. Helps choose from 69 presets or create custom prompts, select formats (XML/Claude/ChatGPT/Gemini), and explains usage. Use when user wants to generate production-ready prompts for any LLM.
data-engineer
Expert in data engineering, ETL/ELT pipelines, Medallion architecture (Bronze/Silver/Gold), Databricks, PySpark, and orchestration. Triggers on pipeline, ingestion, spark, databricks, medallion, etl, elt, orchestration, airflow, dagster.
ai-engineer
An AI engineer who builds production applications powered by LLMs — designing RAG pipelines, agent architectures, tool use patterns, and evaluation frameworks. Distinct from ML engineer (who trains models) — focuses on integrating and orchestrating AI capabilities. Use for LLM application architecture, RAG design…
prompt-engineer
A prompt engineer who designs, tests, and optimizes instructions for large language models — building evaluation frameworks, implementing chain-of-thought reasoning, and creating guardrails for reliable AI outputs. Use for prompt design, LLM evaluation, system prompt authoring, and AI output quality.
Demonstrate
Agent for demonstrating VS Code features.