# Jevable: an MIT agent skill for runnable workflows with human approval **URL:** \ **Category:** Built for Cursor\ **Tags:** skills\ **Created:** [November 10, 2026, 4:10am UTC](https://forum.cursor.com/t/jevable-an-mit-agent-skill-for-runnable-workflows-with-human-approval/177901 "2026-11-10T04:10:18Z")\ **Posts on this page:** 1\ **Page:** 1
**Author:** ![somethingwentwell](https://sea3.discourse-cdn.com/cursor1/user_avatar/forum.cursor.com/somethingwentwell/32/125759_2.png) [@somethingwentwell](https://forum.cursor.com/u/somethingwentwell)\ **Post date:** [November 10, 2026, 4:10am UTC](https://forum.cursor.com/t/jevable-an-mit-agent-skill-for-runnable-workflows-with-human-approval/177901/1 "2026-11-10T04:10:18Z")
Hi, I’m sharing Jevable, an early MIT-licensed agent skill for Cursor and other agents that can use skills and run Node. It helps turn a described process into a workflow app you can inspect, test, and run independently, with explicit human-approval steps. The shipped `request-review` example needs no model account: - `needsReview: false` completes with exit 0. - `needsReview: true` waits at Human Approval with exit 3; nothing is automatically approved. `needsReview` is an explicit boolean, not content-sensitivity detection. ```bash git clone https://github.com/somethingwentwell/jevable-skill cd jevable-skill npm ci --omit=dev ``` Requires Node.js 22.19+, npm, and git. The checkout includes prebuilt `dist/`. This is `0.1.0-alpha.1`, not production-ready. Hosted model providers remain unverified in the public package; custom code nodes validate but do not execute; some test helpers and boundary tooling were reconstructed. [Repository and walkthrough](https://github.com/somethingwentwell/jevable-skill) Feedback on the clone/install path and whether the approval wait is clear would be especially useful. Disclosure: this post was prepared with AI assistance.