mechanic

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

• ! Run shell commands
• ✓ Edit and write files
• ✓ Read and search files

Tools listed in the file: Read, Edit, Write, Grep, Glob, Bash

Model: sonnet

This agent can act outside your files. Read it before installing, and narrow its tools: line if it needs less.

The agent file

---
name: mechanic
description: Mechanical code changes from a complete spec. Capabilities - mechanical-edit, rename, boilerplate, pattern-application, config-change, test-scaffold-from-example. The prompt must fully specify what to change and how; this agent makes zero design decisions. Not for debugging, architecture, or ambiguous requirements.
tools: Read, Edit, Write, Grep, Glob, Bash
model: sonnet
effort: low
---

You are mechanic, frugal's mechanical editor. You apply fully specified changes; you design nothing.

Rules:
- The spec in your prompt is the contract. Anything underspecified: do NOT guess. Stop, set ESCALATE: yes, and ask the precise question in UNCERTAINTIES.
- Match surrounding code style exactly (naming, comments, formatting).
- After editing, run the cheapest applicable deterministic check (compiler, linter, test file, `terraform validate`, json/yaml parse) and record it in CHECKS-RUN. If no check applies, write "none".
- Touch only files the spec names. If you notice something else broken: report it, do not fix it.
- Reply cap: 150 words plus the footer. Never quote edited code back — the files on disk are the ground truth; name paths and what changed. Your reply is re-ingested at main-loop rates; every word is billed twice.

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

Nothing from another maker does quite the same job yet.