Skip to content

Construct Auto Classifier

Shell gate for coding assistants that judges commands by effect and blast radius, asking Jev twelve questions or a chat model, with fast rules first.

Stars
6
Language
TypeScript
License
APACHE-2.0
Published by
Community
Author
godspede
Last verified
Oct 2026

About this project

construct-auto-classifier sits in front of your AI coding assistants' shell execution tools (such as runcommand in Antigravity or bash in OpenCode). It intercepts proposed commands and determines whether they are safe to run automatically, need an agent self-defense explanation, or must escalate to human operator review. Unlike naive keyword filters that blindly reject sudo or /etc, construct-auto-classifier evaluates what the command actually does — its reversibility and blast radius.

Quoted from the project's README. Read the full README.

Highlights

  • Intercepts shell proposals and sorts them into auto-run, explain, or escalate for human review.
  • Default Jev path sends command plus script as state with one allow or deny choice and eleven per-harm risks.
  • Denies on missing answers, low confidence, high risk, or failed calls; fast checks run before the model.
  • Supports agy, OpenCode, and external harnesses while keeping the TypeSafe key out of the gate process.

Watch out

A command is allowed only at 0.6 choice confidence with every risk below 0.7, and build or test allow-rules judge shell verbs rather than code a test runner executes. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

15GitHub stars
Node CLI that lints a file tree against fuzzy natural-language rules in one Markdown file, judging each matched file with a swappable TypeSafe-backed judge.
Jev Agents & MCP#community#jevbest#catalog-import
20GitHub stars
TypeScript recipe library with 248 focused functions that call Jev for routing, reranking, gating and labelling, returning structured ready-or-review decisions for Node backends.
Jev Agents & MCP#community#awesomejev#catalog-import
2GitHub stars
Filters long command output for coding agents against a task description, keeping retained lines unchanged and in order, collapsing dropped runs into markers with a saved copy you can print back later.
Jev Agents & MCP#community#awesomejev#catalog-import
3GitHub stars
Hermes plugin that puts an agent's plans, drafts and claimed completions up for review — plan_review, evidence_review, action_verify — plus local approval rules that make no model request.
Jev Agents & MCP#community#awesomejev#catalog-import
2GitHub stars
A runtime authorization gateway for agent tool calls: deterministic policy plus six narrow Jev questions produce evidence, then a single-use grant bound to the exact approved action is consumed at execution.
Jev Agents & MCP#community#awesomejev#catalog-import
1GitHub stars
A Claude Code plugin that verifies every tool call before it runs, asking Jev five parallel questions about the plan, scope, prior action and stated intent, then allowing, asking or blocking. Fresh installs ship in audit mode.
Jev Agents & MCP#community#jevbest#catalog-import
Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai