gutcheck
github.com/sfmqrb/gutcheck- Category
- Developer Tools
- Rank
- No. 3613Tools index
Previous survey · No. 3469 ·
- Pricing
- Open Source
- Type
- TOOL
- Use case
- Data Processing
- Interfaces
- CLI
- GitHub
- 14 stars
- Latest release
- v0.5.0
- Date
About
A single Rust binary that filters, scores, or classifies text lines against a plain-English question, using a local CPU decision model (Laya, run via ONNX) with no API key, Python runtime, or server required. It can keep lines matching a description, print a match probability per line, or sort input into custom labels.
At a glance
gutcheck filters lines or records by a plain-English question using a local model. It is worth trying when the thing you want to find has several phrasings and a fixed pattern keeps missing it.
Good fit
Developers sorting synthetic or approved logs, issue reports and structured records at a terminal.
Consider something else when
Use a deterministic rule when missing a match would have security or compliance consequences.
Start here
Label a dozen sample lines, including ambiguous examples. Compare the matches at two thresholds and check which misses would matter.
Cost and requirements
Runs inference on the CPU and uses no API key. The project documents a 1.3 GB model download on first run.
Limits to know
The match score depends on the model and question. A clean result is insufficient evidence that a file contains no secrets or other sensitive material.
Sources checked 2026-11-05. Source-based assessment.
Why it made the leaderboard
Lets developers pipe logs or text through a local, offline model to filter or label lines by meaning (e.g. 'is this an error?') the same way they'd use grep, without an API key or network call.
Tags
Tech Stack
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.