User repository on GitHub
Automatically label GitHub Issues and Pull Requests using lightweight decision models like Jev and Microsoft-Decision-1.
- Forks
- 1
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Integrations & Adapters
DeciLabel
A Node 24 GitHub Action that classifies issues and pull requests with a typed Decisions API and adds only the labels the available evidence supports. It can call Jev through TypeSafe, OpenRouter (the default) or OpenAI's gpt-6-luna, and never removes, replaces or creates labels.
- Stars
- 1
- Language
- TypeScript
- License
- MIT
- Reading time
- 3 min
- Published by
- Community
- Author
- evanprogramming
- Last verified
- Oct 2026
Works forJev for Classification
Used inJev for Software & DevTools
- Languages
- TypeScript
- Capabilities
- Open sourceRequires API key
Highlights
- Three providers: OpenRouter by default, TypeSafe direct with model typesafe/jev-1.13, or OpenAI's public-beta Decisions API with gpt-6-luna.
- It adds only labels the evidence supports and never removes, replaces or creates them; the run summary explains which labels were selected, added or rejected.
- Watches issues opened, edited and reopened plus pull requests opened, edited, synchronize, reopened and ready_for_review.
- Ships as a Node 24 GitHub Action with dependencies included in dist/index.js; the minimal example pins EvanProgramming/[email protected].
- Because it uses pull_request_target, public repositories must allow its events in Actions policies; GitHub's announced enforcement date is November 2, 2026.
Watch out
Grounded on the truncated README and GitHub metadata read on 2026-10-15; the capture ends mid-paragraph in the Models and decisions section, so the full provider comparison it was describing is not reproduced. No classification accuracy is quoted anywhere in the capture — the action's stated guarantee concerns which labels it may add, not how well it classifies.
Questions about this entry
Short answers drawn from the linked project or page.
A GitHub Action that classifies issues and pull requests with a typed Decisions API and adds only the labels supported by the available evidence. It listens to new and updated issues and PRs, and it never removes, replaces or creates labels.
OpenRouter by default, Jev through TypeSafe's direct API with provider: typesafe and model typesafe/jev-1.13, or OpenAI's Decisions API with gpt-6-luna. Microsoft's model also works through OpenRouter as microsoft/microsoft-decision-1.
Add OPENROUTER_API_KEY, TYPESAFE_API_KEY or OPENAI_API_KEY as a repository secret, copy the minimal workflow into .github/workflows/decilabel.yml, make sure the labels you want already exist in the repository, and commit. The action then runs on issue and pull-request events.
No. It selects from labels that already exist in the target repository and adds only the ones the evidence supports. If a label you expect is missing, create it first — the run summary explains which labels were selected, added or rejected.
For issues: opened, edited and reopened. For pull requests: opened, edited, synchronize, reopened and ready_for_review. The minimal example requests only the label-write permissions it needs, and reviews who may trigger it on public repositories.
GitHub is rolling out workflow execution protections that can stop pull_request_target from starting, with an announced enforcement date of November 2, 2026. A policy-blocked workflow never starts, so check Actions policy insights and the repository or organization policy configuration.
With provider: typesafe the action calls POST https://api.typesafe.ai/v1/systemone using TypeSafe's jev-latest alias, and that provider supports Jev only. OpenRouter forwards model IDs to its decisions endpoint and lets the endpoint validate which are supported; the OpenAI path uses the public-beta Decisions API and gpt-6-luna.
More Jev resources like this
A GitHub App that routes pull-request review work through Jev: it ingests the webhook, normalizes the PR context, gets a LOW, NORMAL or SPECIALIST decision, and a deterministic policy layer maps it to auto-approval, a review request or a check-run update.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
hexyedev/jevprPython
hexyedev/jevprPythonA GitHub Action that asks Jev one yes/no question per thing worth checking on a pull request and approves only when every answer clears your confidence threshold; otherwise it skips and comments with the numbers.
Jev Integrations & Adapters#community#catalog-import#logicrw
A Claude Code plugin that picks the model tier and thinking effort for every turn with Jev and weighs each switch against what it costs, since moving a long conversation to another model rewrites its prompt cache and can cost more than it saves.
Jev Integrations & Adapters#community#catalog-import#logicrw
A Google Sheets add-on that classifies, tags and scores text with plain-language formulas: JEV_IF, JEV_PROB, JEV_CHOICE and JEV_SCORE return typed values, and uncertain rows can say UNSURE instead of guessing.
Jev Integrations & Adapters#community#catalog-import#cobanov
Community
cab14bacc/jev-sheetsTypeScript
cab14bacc/jev-sheetsTypeScriptA Claude Code mod that adds $.jev to the engine interface, letting plugin hooks ask Jev for Choice, Score and Noul answers over one state via TypeSafe, OpenRouter, Vercel, Cloudflare, LiteLLM or a custom endpoint.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
chrishan17/claude-jev-modTypeScript
chrishan17/claude-jev-modTypeScriptA pi coding-agent extension that compacts long transcripts without rewriting them: Jev answers only keep/kind questions about chunked history, original lines go into a verbatim ledger, and compaction is a deterministic render plus durable topic files. Calls the TypeSafe System One endpoint directly.
Jev Integrations & Adapters#community#catalog-import#logicrw
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.