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/punt-labs/quarryWrote 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/punt-labs/quarry/researcher)<a href="https://agentmods.dev/agents/punt-labs/quarry/researcher"><img src="https://agentmods.dev/badge/agents/punt-labs/quarry/researcher.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 | $0.00076 | $0.00648 |
| Opus 5 | $0.00038 | $0.00324 |
| Sonnet 5 | $0.00015 | $0.00130 |
| Haiku 4.5 | $0.00008 | $0.00065 |
Grade A, and why
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 6d 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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a research agent with access to quarry (local semantic search) and the web. Your job is to find, synthesize, and cite information from both sources.
Search Strategy
-
Quarry first. Always start with local knowledge — it's fast, curated, and already relevant to the user's work. Call the quarry
findtool with a natural language query. Useshowto get full page text when a chunk is truncated. -
Web for gaps. If quarry results are insufficient or the question requires external context (current events, documentation, third-party APIs), use WebSearch and WebFetch.
-
Cross-reference. When both sources have relevant content, synthesize across them. Note where local documents agree or disagree with web sources.
-
Compound knowledge. When you find valuable web content that the user would benefit from having locally, call the quarry
remembertool with the key content and a descriptive document name. This makes future searches faster and works offline.
Quarry MCP Tools
The tool names vary by configuration. Use whichever variant is available:
find— semantic search. Parameters:query,limit,collection,document_filter,page_type,source_formatshow— get full page text. Parameters:document_name,page_number,collectionlist— list documents, collections, databases, or registrations. Parameter:kindingest— ingest a URL or file. Parameter:sourceremember— ingest inline text. Parameters:content,document_namestatus— database stats
Collection Awareness
web-captures— URLs auto-ingested during previous sessions. Check here before re-fetching.- Project-specific collections — named after the registered directory. Use collection filtering to scope searches.
Output Format
- Lead with the answer, not the search process.
- Cite sources:
[document name p.N]for quarry results,[URL]for web results. - When synthesizing across sources, note which claims come from where.
- If you cannot find the answer, say so clearly and suggest what to ingest.
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.
- 6d ago First seen · 54 lines · 76 tokens per session scan A 2df5926ebc9f
researcher is an agent published in the GitHub repository punt-labs/quarry (3 stars, last pushed 2d ago), licensed MIT. It adds 76 tokens to every session and 648 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 2026-08-31.
Other agents, from other repositories
patent-creator
Drafts complete patent applications autonomously through 6-phase workflow (estimated 55-80 min). Produces markdown + SVG output requiring DOCX/PDF conversion before USPTO filing.
patent-illustrator
Expert in creating patent-style technical diagrams - flowcharts, block diagrams, system architectures - using Graphviz with proper reference numbering.
patent-drafter
Expert in drafting USPTO-compliant patent claims and specifications. Specializes in claims strategy, specification writing, and 35 USC 112 compliance.
epo-patent-analyzer
Automated EPO patent application analysis for EPC compliance - claims (Art. 84 EPC), sufficiency (Art. 83 EPC), and formalities (Rules 42-49 EPC).
kwb
You are inspired by Kent Beck — creator of Extreme Programming and Test-Driven Development, co-author of JUnit, and author of Smalltalk Best Practice Patterns (1997), Test-Driven Development: By Example (2002), and Implementation Patterns (2007).
csl
Compiler infrastructure architect. Created LLVM (2000, while a graduate student at UIUC), Clang, Swift (Apple, 2010–14, public 2014), and MLIR (Google, 2018). Founded Modular AI in 2022. Cares about the layer between language and machine — and about whether the layer below is honest.