Jev directory
CommunityJev Tools & CLIs
Aiscript
Python CLI that turns a plain-English request into a cached shell or Python script and reviews each generated script before it is saved. Risky scripts need explicit approval, with an optional decision-model reviewer that can be Jev, OpenAI Decisions, or a local Ollaya model.
- Stars
- 0
- Reading time
- 2 min
- Published by
- Community
- Added
- Oct 2026
Highlights
- Installs with uv tool install, pipx install, or pip install; requires Python 3.10 or newer.
- Generated scripts are cached, so later runs need no network and no configuration.
- A built-in static regex scan flags privilege escalation, recursive file changes, and curl | sh.
- The safety review fails closed: an unreachable or erroring reviewer treats the script as risky.
- Non-interactive stdin refuses risky scripts, and stdout plus the exit code belong to the script.
Questions about this entry
Short answers drawn from the linked project or page.
Install with uv tool install aiscript, pipx install aiscript, or pip install aiscript. Python 3.10 or newer is required. Upgrade with the same tool you installed with; config, prompts, and cached scripts in ~/.config/aiscript/ are kept.
An OpenAI key for the provider that writes the scripts and TYPESAFE_API_KEY for Jev review, placed in ~/.config/aiscript/config.yaml or exported as OPENAI_API_KEY and TYPESAFE_API_KEY. Placeholder keys such as "..." are ignored, so environment variables still work.
Yes. Set reviewer_provider to ollaya for a local, keyless decision model, with jevk5:4b as the default model on localhost:11435. Comment out reviewer_provider to rely only on the built-in static scan.
The review fails closed: the script is treated as risky. You see the reasons and the code and must answer [y/N]; declining saves nothing. With a non-interactive stdin, risky scripts are refused outright.
No. The static scan plus decision model are described as a best-effort safety net, not a sandbox, so you should read what you approve before it is saved or executed.
Everything lives in ~/.config/aiscript/, including config.yaml, which is written in mode 600 because it can hold API keys. The documentation advises keeping that file out of git.
Everything after the command name is passed untouched to the script, including --help. aiscript messages go to stderr, stdout belongs to the script, and the exit code is the script's own.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
YouTube: Kubernetes admission control with a decision model, fail-closed.
Hacker News: SafeDep Gryph plus Jev checks for compromised coding-agent behavior.
2 pointsDetecting Compromised AI Coding Agents with Jev and GryphRead the thread on Hacker NewsX: Jev and Laya on security queries
X: Testing Jev for prompt injection
More Jev resources like this
A CLI that turns Jev into Unix pipeline primitives: is, choose, score, filter, and guard commands return typed verdicts, stable exit codes, and versioned JSON without prompt or parsing glue.
Jev Tools & CLIs#community#python#cli
Community
sharziki/semdecidePython
sharziki/semdecidePythonA pre-commit hook that answers five Jev questions about the commit message and staged diff together, warning on mismatches and blocking only on a credential-shaped line.
Jev Tools & CLIs#community#python#coding-agent
Community
valentynkit/jev-commitPython
valentynkit/jev-commitPythonA Node CLI that sends source, configuration, build, and CI files to Jev and reports suspicious files and lines with probabilities, as a second opinion before running unfamiliar code.
Jev Tools & CLIs#community#typescript#cli
Community
luantak/is-maliciousTypeScript
luantak/is-maliciousTypeScriptAgent-ergonomic CLI following the AXI conventions that gives coding agents Jev judgments for blocking risky tool calls, screening fetched content for prompt injection, triaging logs, flagging risky diffs, and ranking many items. Routine commands are decided locally at no cost.
Jev Tools & CLIs#community#typescript#cli
Community
shiftynick/jev-axiTypeScript
shiftynick/jev-axiTypeScriptA local, non-autoregressive security guardrail suite powered by Convai Innovations' Laya model, returning typed decision probabilities for security inspection. It positions itself against Jev on latency and against generative LLMs on hallucination risk.
Jev Tools & CLIs#community#github#security
A Hermes Agent plugin that replaces the auxiliary approval reviewer with TypeSafe's Jev decision model: one request asks six typed questions and deterministic code returns APPROVE, DENY, or ESCALATE.
Jev Tools & CLIs#community#python#agents
Community
anpicasso/hermes-jev-approvalsPython
anpicasso/hermes-jev-approvalsPythonFrom 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.