User repository on GitHub
jev: a Jev (TypeSafe System One) toolkit CLI for coding agents — read less, decide with calibrated confidence. Unofficial.
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Tools & CLIs
jevkit (Go)
An unofficial Go CLI that puts Jev in a coding agent's toolbox: the agent can ask where an answer lives and read only those lines, or hand Jev a yes/no or multiple-choice call and get a calibrated verdict tuned to the cost of a wrong call.
- Stars
- 3
- Language
- Go
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- bragamat
- Last verified
- Oct 2026
Used inJev for Software & DevTools
Highlights
- Go CLI for coding agents with two moves: read less by asking Jev where the answer is, and decide with calibrated confidence.
- Verdicts come back as ACT, CONFIRM or REPHRASE, tuned to how costly a wrong call would be.
- README states Jev answers typed questions with probabilities in a few hundred milliseconds at $0.042 per million input tokens.
- The gateway, context-diet and pruning experiments were removed; docs/STUDIES.md records what they measured.
- Unofficial community project, not affiliated with TypeSafe AI; needs your own key from console.typesafe.ai. MIT, Go.
Watch out
README and GitHub metadata reviewed 2026-10-17; the excerpt truncates right after the Install heading, so install commands, flags and the STUDIES.md findings are not evidenced. Created 2026-10-05, last pushed 2026-10-07.
Questions about this entry
Short answers drawn from the linked project or page.
A Go command-line toolkit, installed as jev, that puts Jev, TypeSafe's System One model, in a coding agent's toolbox so the agent can look up locations and make calibrated yes/no or multiple-choice calls instead of re-reading context.
Instead of pasting a whole file into the conversation, where it would be paid for again on every later turn, the agent asks Jev where the answer is and reads only those lines, per the README's framing.
They are the verdicts Jev hands back for a yes/no or multiple-choice call about the agent's own work, tuned to how costly a wrong call would be, per the README's description of the decide move.
The README states Jev answers typed questions with probabilities in a few hundred milliseconds at $0.042 per million input tokens, so a lookup costs a fraction of pasting the file. You need your own TypeSafe API key.
No. The README carries an explicit notice that jevkit is a community project, not affiliated with, endorsed by or sponsored by TypeSafe AI, and that Jev and TypeSafe are their names.
They were removed. The README says jevkit is now only the CLI an agent can call as a tool, and that docs/STUDIES.md records what the gateway, context-diet and pruning experiments measured.
MIT-licensed per GitHub metadata, with 3 stars, created 2026-10-05 and last pushed 2026-10-07. The digest excerpt truncates after the Install heading, so binary installation steps are not evidenced here.
Other listings of this project
Forks and copies of this project in the directory. This entry is the canonical one.
A Rust CLI that sends terse YAML or JSON question sets to Jev, validates them offline with 13 lint rules before any billed call, and prints parsed answers as JSON.
Jev Tools & CLIs#community#rust#cli
Community
ariel-frischer/jevkitRust
ariel-frischer/jevkitRustSelf-hosted Jev tools for coding agents: one stdlib HTTP server holds the TypeSafe key while CLIs (ask-jev, jev-review), agent skills and a Pi tool ask Jev typed questions about code instead of reading it into agent context.
Jev Tools & CLIs#community#catalog-import#logicrw
Community
maxivanyshen/jevkitPython
maxivanyshen/jevkitPythonMore Jev resources like this
Read-only Go CLI that semantically checks your files against your own rules with Jev, locally or in CI, printing text or JSON and caching results between runs.
Jev Tools & CLIs#community#github#go
Community
Alurith/jeffGo
Alurith/jeffGoA single-binary gate for Impatient Programming that fails a pull request when an experienced reviewer would ask for the change to be simplified: it asks Jev about each changed file and tells the coding agent what to fix, with no LLM or parser inside.
Jev Tools & CLIs#community#github#go
Community
tools4imps/qualmGo
tools4imps/qualmGoA Go CLI by dorkitude that skims pages, files, and lists with Jev, keeping only the chunks that help with your stated goal. The README's example kept 4 of 40 chunks and cut 71k characters to 7.6k, an 89% saving, for whatever agent reads next.
Jev Tools & CLIs#community#github#go
Community
dorkitude/skimmerGo
dorkitude/skimmerGoAn open-source merge gateway for repositories where agents write most commits: a System One model answers five fixed questions per pull request with calibrated probabilities in roughly 300 ms, a YAML policy turns them into pass, hold or block, and a planner lands non-conflicting PRs in waves.
Jev Tools & CLIs#community#github#go
Community
haalp/mergeonePython
haalp/mergeonePythonGo CLI for Jev with noul, choice, score, eval, rank, and models commands: rank sends one Score question per item per weighted YAML dimension in a single request and sums weight times score in code, while threshold flags turn a failed gate into exit code 10 for scripts and CI.
Jev Tools & CLIs#community#cli#go
Community
saembit/jeff-cliGo
saembit/jeff-cliGoVercel Labs' Unix-like coding agent CLI written in Zig, shipping a typesafe_permission_reviewer builtin that runs the agent's permission decisions through Jev rather than an LLM call. Model-agnostic and embeddable via libfx and ACP.
Jev Tools & CLIs#community#coding-agents#cli
Community
vercel-labs/fxZig
vercel-labs/fxZigFrom 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.