Jev application
Build and run agents you can see, understand and trust.
Details →★ 32,297
feature-spec
Agent skill for multi-doc feature specs that can't contradict themselves — or quietly ship a claim nobody verified. One facts registry, mechanical audits, and a verify mode that gates 'shipped' on evidence.
Project facts
- Relationship to Jev
- Jev application
- Evidence
- Documented
- Language
- Python
- License
- MIT
- Origin
- Original repository
- Repository status
- Not archived
- Created
- 2026-08-19
- GitHub stars
- 1
- Evidence checked
- 2026-09-24T08:40:10.508Z
- Metadata checked
- 2026-09-24T08:40:10.508Z
- Check status
- current
Stars measure the whole repository, including work unrelated to Jev.
Evidence and scope
Documented records the linked documentation or source. JevHunt has not independently run or benchmarked this project.
It asks one bounded question per candidate — is this the real thing, or one of the exempt shapes the protocol names in prose — and sorts them into `keep`, `uncertain` and `suppressed`, so the ones worth a document read come first; suppressed candidates are still printed, because a list nobody is told to read is the failure the human pass exists to close. It writes no finding and assigns no severity. Ranking needs the `typesafe-sdk` package and an API key, and having neither is a supported state
Evidence commit: 6d581a739503ec3f5a86fbda93162926f451ee90
Discovered through: github-search.
How we review →