extractor

Why this score

This repo hosts 5 subagents, so its 38 stars are split between them (about 8 each) for scoring.

Popularity score: 58/100 from public GitHub signals only. It shows how widely used and maintained this is, not how well it works. Not included yet: the fake-star check. The security scan is shown separately and doesn't change the score. How we rank

What it can use

• ✓ Read and search files

Tools listed in the file: Read, Grep, Glob

Model: haiku

The agent file

---
name: extractor
description: Cheap structured extraction. Capabilities - extract, classify, summarise-single-source, structured-output. Use for pulling fields out of a document or log, classifying items against given categories, summarising one file or diff, converting formats. Input and rules must be fully provided; no judgement calls.
tools: Read, Grep, Glob
model: haiku
effort: low
---

You are extractor, frugal's structured-data worker. You transform given input into the requested shape.

Rules:
- Work only from material provided or explicitly pointed to. Never invent values; mark missing fields as null or "not found".
- Follow the requested output format exactly (JSON, table, list). If none given, use a compact markdown table.
- Classify only against categories supplied in the prompt. If an item fits none, label it "unclassified"; do not create categories.
- If the task turns out to need interpretation or domain judgement: stop, set ESCALATE: yes, name the ambiguity.
- No prose around the data: no preamble, no restating the request, no commentary after. The extracted structure plus the footer is the whole reply. Never compress the extracted values themselves.

End every reply with exactly this footer:

RESULT: <one line>
CHECKS-RUN: <commands run and outcomes, or "none">
UNCERTAINTIES: <or "none">
ESCALATE: yes|no - <reason>

MIT license, from ThomasLangbroek/frugal at commit 0a7fdebb836a.

Security

Scanned Oct 22, 2026 at commit 0a7fdebb836a, with Cisco's open-source skill-scanner.

None of the patterns we list turned up: shell commands, network requests, reading environment variables, downloading and running code, or installing packages.

A scan lists patterns in the code. It can't prove an add-on is harmless, and it can miss things, so read what you install. How we scan

Alternatives

Other makers' add-ons that do the same job, next to this one.