User repository on GitHub
Local Codex tool-output filtering with OpenAI Decisions by default and optional TypeSafe Jev support.
- Forks
- 1
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
Jev directory
CommunityJev Integrations & Adapters
Codex Decision
A local Codex extension that filters repetitive lines from supported tool output with OpenAI Decisions by default or TypeSafe Jev optionally, keeping errors visible and linking every shortened result to the full original.
- Stars
- 2
- Language
- Rust
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- philippelhaus
- Last verified
- Oct 2026
Highlights
- Installs as a single VSIX — code --install-extension codex-decision-0.11.8.vsix --force — for Windows VS Code with a Linux x86_64 WSL environment.
- Connect Decision asks you to sign in to Codex, pick OpenAI Decisions or TypeSafe Jev, and test that provider's key; the key stays out of chat.
- Monitor keeps full output and Filter shortens approved output; the relevance cutoff, API key and log retention live in Codex settings under Decision.
- The Decision panel beside Terminal shows blue rows for proposed omissions and gold rows for kept lines, with relevance bars and hover excerpts.
- The README warns removal counts do not establish downstream token savings because code-mode scripts can still serialize the original result.
Quickstart
bash
code --install-extension codex-decision-0.11.8.vsix --forceWatch out
Copy reviewed against the truncated repo README on 2026-10-15; screenshots use synthetic data and the excerpt cuts off in the documentation links. The README's own investigation says the panel's removal counts do not prove downstream token savings, and version 0.11.8 is the release named in the excerpt.
Questions about this entry
Short answers drawn from the linked project or page.
A local filtering layer for Codex tool output: it selects repetitive lines for removal from supported output while errors and useful evidence stay visible, and every shortened result links to the complete original. The README names version 0.11.8.
Two: OpenAI Decisions (the default) and TypeSafe Jev as an option. You sign in to Codex, select a provider in Connect Decision, enter and test that provider's API key, and each thread remembers its own on/off choice.
Install the single VSIX on Windows VS Code with a Linux x86_64 WSL environment, run Developer: Reload Window after the integration repair, then open your Codex thread. Review the hook in /hooks if Codex asks for trust; no second package is needed.
Monitor keeps the full output while Filter shortens approved output. Both share settings for the relevance cutoff, API key and log retention in Codex settings under Decision, and the settings apply to all sessions.
Open the Decision panel beside Terminal and Output: blue rows mark proposed omissions, gold rows mark kept lines, and bars show relevance with hover excerpts and reasons. Latest shows the current result and Totals shows saved thread counts.
The README is explicit that it is not established: code-mode scripts can still serialize the original tool result after a hook records a replacement, so the panel's removal counts do not establish downstream token savings. An example panel shows 47 of 50 lines removed.
Unsupported builds produce a visible diagnostic. Decision: Check Integration reports status and Decision: Repair Integration restores the controls and reopens a hidden panel, according to the README's setup documentation.
More Jev resources like this
A GitHub App that routes pull-request review work through Jev: it ingests the webhook, normalizes the PR context, gets a LOW, NORMAL or SPECIALIST decision, and a deterministic policy layer maps it to auto-approval, a review request or a check-run update.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
hexyedev/jevprPython
hexyedev/jevprPythonA Claude Code mod that adds $.jev to the engine interface, letting plugin hooks ask Jev for Choice, Score and Noul answers over one state via TypeSafe, OpenRouter, Vercel, Cloudflare, LiteLLM or a custom endpoint.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
chrishan17/claude-jev-modTypeScript
chrishan17/claude-jev-modTypeScriptA Python library that runs LLM and Jev calls as Polars expressions: parallel inference across providers, Pydantic-validated outputs as struct columns, and a native TypeSafe System One layer for per-row decision questions.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
pnthn-ai/polar_llamaPython
pnthn-ai/polar_llamaPythonA tool-permissions layer for the Vercel AI SDK backed by Jev through the Vercel AI Gateway: a pre-flight approval gate that decides auto-approve, ask-human or block, plus tool-list pruning that drops irrelevant definitions to save context tokens.
Jev Integrations & Adapters#community#catalog-import#logicrw
A governed execution layer and drop-in proxy between typed-decision models like Jev and callers such as Claude Code, Cursor/MCP and SDKs: raw probabilities become deterministic policy actions from ALLOW to DENY with a DuckDB audit trail.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
0xshikhar/jev-fusePython
0xshikhar/jev-fusePythonA Claude Code plugin that adds five Jev tools — jev_choice, jev_noul, jev_score, jev_ask and jev_doctor — so small decisions like safety checks, tool choices and review verdicts come back as typed answers with calibrated confidence, labelled auto or escalate.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
darwintechlab/claude-jevJavaScript
darwintechlab/claude-jevJavaScriptFrom 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.