yutkat/github-star-organizer-jev
User repository on GitHub
Python tool that classifies GitHub stars into existing GitHub Lists using TypeSafe Jev
- Forks
- 0
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Tools & CLIs
Github Star Organizer Jev
A Python CLI that sorts GitHub stars into your existing GitHub Lists using README excerpts and repo metadata, writes a JSON classification report, and applies it only when you ask.
- Stars
- 1
- Language
- Python
- License
- MIT
- Published by
- Community
- Author
- yutkat
- Last verified
- Oct 2026
About this project
Use TypeSafe Jev to classify GitHub stars against your existing GitHub Lists. Generate a JSON classification report and optionally apply its assignments to GitHub. Install Python 3.11 or later and uv. Create at least one List at . Set STARLISTSTOKEN to a personal access token owned by the account whose stars and Lists you want to classify. A classic PAT with the user scope can be used for Lists; private repositories also require repository access (repo for a classic PAT).
Quoted from the project's README. Read the full README.
Works forJev for Classification
Used inJev for Software & DevTools
- Languages
- Python
Highlights
- Each repository sends its name, description, topics, language, other metadata and the first 2,000 README characters with the List names.
- Apply is a separate command that dry-runs first, checks the token owner and that destination Lists still exist, then adds memberships.
- The report records the suggested List, confidence, probabilities, model, token usage and request time including retries.
- Up to 254 existing Lists are supported, leaving one choice reserved for the No matching category escape hatch.
Watch out
The README sent to the model has image and badge markup, HTML comments and tags, and empty lines stripped, and it is not cached, so a re-run re-fetches. If a README is missing or unreadable, classification falls back to metadata alone, which is a materially thinner signal. Confidence describes the probability distribution and is explicitly not a measured classification accuracy. Applying touches only the token owner's existing Lists; Lists are never created or deleted, No matching category results are skipped, repositories already in the target List are skipped, and a rerun skips work that already succeeded. Membership reads must be complete - a failed or partial read stops application - and GitHub's update replaces an item's entire List membership set, so concurrent edits while applying are a real hazard. In GitHub Actions, both the apply step and the weekly schedule are commented out until you uncomment them. starred_total is GitHub's reported total even when --limit sampled fewer. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
Not by default. Classification only writes a JSON report - to --output or stdout - and GitHub Actions runs only generate reports until you uncomment the Apply List assignments step. Applying is a separate github-star-organizer-jev-apply command that takes --report, supports --dry-run to validate against current GitHub data without changes, and needs no Jev credentials. Lists are never created or deleted.
A classic PAT owned by the account whose stars and Lists you are classifying, with the user scope for Lists; private repositories also need repository access (repo for a classic PAT). Only the token owner's existing Lists can be used. The README notes GitHub's update replaces an item's entire List membership set, so avoid concurrent membership edits while applying, and a failed or partial membership read stops application rather than half-applying it.
Accuracy is not claimed: the README says confidence describes the probability distribution and is not a measured classification accuracy, and there is no evaluation set. You can try --limit 10 for a small sample first. --endpoint or JEV_ENDPOINT selects another Jev-compatible provider that must accept the same typed-question request and response format with Bearer auth; the CLI flag beats the env var, the default is https://api.typesafe.ai/v1/systemone, and --model defaults to jev-latest.
For each repository it fetches the default branch's README through the REST API, strips image and badge markup, HTML comments and tags and empty lines, and sends the first 2,000 characters alongside the name, description, topics, language, other metadata and the List names and descriptions. READMEs are not cached, and a repo without one is classified from metadata alone.
Python 3.11 or later with uv. Use --limit 10 for a small sample; a limit of 0, the default, means all stars.
It only generates classification reports. The apply step and the weekly schedule are both commented out, and runs are serialized with a six-hour timeout and 30-day artifact retention.
It validates the report, checks the token belongs to its owner, and confirms destination Lists still exist. It adds the selected List, skips repositories already in it, skips no-matching-category results, never creates or deletes Lists, and a rerun skips already-succeeded assignments.
More Jev resources like this
44GitHub stars
Mac app that turns one short English or Japanese sentence into Ableton Live mixer, transport, clip, note, device, and track edits from a pop-up bar.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
okinaaudio/live-jevPython
okinaaudio/live-jevPython547GitHub stars
Live SEO audit that crawls a site from one homepage URL, checks 52 rules tied to Google Search Central, and asks Jev typed questions about every page to rank fixes.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
AgriciDaniel/jev-seoPython
AgriciDaniel/jev-seoPython27GitHub stars
Local-first trading journal and review harness that is read-only over markets, captures runs into frozen envelopes and evidence packs, and requires explicit human promotion before any rule returns.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
myc0576/SmartMoney-CubPython
myc0576/SmartMoney-CubPython39GitHub stars
Video tool where Jev scores subtitle segments and code assembles a short highlights cut plus a de-watered full cut with timestamped summaries.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
cclank/jevclipPython
cclank/jevclipPython30GitHub stars
Forkable weekday radar that asks Jev one yes-or-no question per interest over each new arXiv announcement and publishes a page plus RSS feed.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
Eliot5566/JEV-Paper-RadarPython
Eliot5566/JEV-Paper-RadarPython38GitHub stars
Async supervisory layer for Hermes Agent that locks Definitions of Done, proves mechanical criteria locally, and calls Jev only for unresolved semantic checks under a token budget.
Jev Tools & CLIs#community#jevbest#catalog-import
Community
keeltrace/hermes-nervePython
keeltrace/hermes-nervePythonFrom the Jev guides
Original write-ups on what this entry is for.
The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.
Jev Tools & CLIs#gateway#vercel#cloudflare
Read article
Taking a decision checkpoint out of Python: what the GGUF and browser WebGPU exports preserve, what they change, and the parity evidence the model cards publish.
Jev Tools & CLIs#open-models#local#onnx
Read article
Jev labels a record that already exists. Code moves, flags, or holds it. How the document, mail, and transfer tools keep the action out of the model.
Jev Tools & CLIs#classification#email#security
Read article
From the Jev community
Posts from builders shipping with Jev right now.