Organization repository on GitHub
Configurable semantic CI checks for repository files using TypeSafe Jev
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Integrations & Adapters
Jev Linter Action
A GitHub Action that reviews repository files with your own yes/no questions: Jev returns probabilities and the check passes only when every expected answer meets its configured threshold.
- Stars
- 1
- Language
- JavaScript
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- sable-inc
- Last verified
- Oct 2026
- Languages
- JavaScript
- Platforms
- API
- Capabilities
- Open sourceRequires API key
Highlights
- Questions are a YAML list inside the workflow; a plain question expects yes at probability 0.8 or higher, or set expect and minProbability.
- glob accepts one pattern or a newline-separated list; per-file: true reviews each file independently instead of all together.
- Missing inputs, malformed questions and ambiguous combinations fail before any API call is made.
- Runs on Node 24 with a checked-in bundle, so consumers need no installation step.
- Optional locate mode asks Jev itself which authored paragraphs contribute to failed rules and maps them back with source-glob.
Watch out
Grounded in the truncated README and GitHub metadata reviewed on 2026-10-15; the README was cut off after the locate example, publishes no benchmarks, warns that split bundles cannot support contradiction or duplication checks, and requires a TYPESAFE_API_KEY secret.
Questions about this entry
Short answers drawn from the linked project or page.
A GitHub Action that reviews repository files with your own yes/no questions. Jev returns probabilities, and the action passes only when every expected answer meets its configured threshold, turning semantic checks into a CI gate.
Questions are a YAML list in the workflow. A plain question expects yes with probability at least 0.8; for negative questions or custom thresholds use the object form with id, question, expect and minProbability, as the README's forced_scope example shows.
A model name such as jev-1.13.0, a glob selecting the files to review, the questions list, and an api-key passed from a secret like TYPESAFE_API_KEY. It runs on Node 24 with a checked-in bundle, so consumers add no install step.
The README warns against using contradiction or semantic-duplication questions as consistency checks: oversized bundles are split, distant chunks are never compared, and source localization cannot recover a missing comparison or establish a contradiction pair.
With locate: true, Jev itself is asked which authored paragraphs contribute to failed rules, and source-glob maps findings back to the original source files. The README notes no generative LLM or second provider is needed; checkout with fetch-depth 0 helps recover patches.
Yes: build the agent first, then point glob at the built artifact, for example acme/demo/.sable/build/agent.bundle.json. The README's example uses Sable's build --bundle output; the action reviews only locally bundled skills, not remote-only platform skills.
Missing inputs, malformed questions and ambiguous combinations fail before any calls are made, so a broken workflow stops early instead of spending Jev requests. glob accepts a single pattern or a newline-separated list, and per-file: true reviews files one by one.
More Jev resources like this
A 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 governed execution layer and drop-in proxy between typed-decision models like Jev and callers such as Claude Code, Cursor/MCP and SDKs: raw probabilities become deterministic policy actions from ALLOW to DENY with a DuckDB audit trail.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
0xshikhar/jev-fusePython
0xshikhar/jev-fusePythonA GitHub Action that treats .jev.yml as a semantic linter for pull requests: each rule is a typed Jev question (Noul, Choice or Score) and the check fails, comments or passes from those numbers. It writes no reviews, suggests no patches and approves nothing.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
thestral12/pr-sieveTypeScript
thestral12/pr-sieveTypeScriptAn opencode plugin that turns an agent's small decisions — is this command safe, which tool fits, is this PR ready — into typed Jev questions with calibrated confidences, labelling each answer auto or escalate so the agent knows when to ask for help.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
darwintechlab/openjevJavaScript
darwintechlab/openjevJavaScriptA 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 Playwright reporter that uses Jev to score which existing E2E tests a code change could affect and runs only those above a confidence threshold before browser execution, failing open to the full suite.
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.