murilogteixeira/opencode-jev-pre-llm
User repository on GitHub
OpenCode V2 plugin that uses TypeSafe JEV to classify prompts before LLM reasoning
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Integrations & Adapters
Opencode Jev Pre LLM
An OpenCode V2 plugin that asks TypeSafe Jev whether a prompt is deterministic before the LLM reasons, injecting a JEV PRE-ANALYSIS block and temperature 0 for deterministic turns so simple prompts skip chain-of-thought. Jev failures fail open and the session continues normally.
- Stars
- 0
- Language
- TypeScript
- License
- MIT
- Reading time
- 2 min
- Published by
- Community
- Author
- murilogteixeira
- Last verified
- Oct 2026
- Languages
- TypeScript
- Capabilities
- Open sourceRequires API key
Highlights
- Hooks OpenCode's prompt and context events; one Jev classification per turn, later steps reuse a cached verdict with a TTL defaulting to 10 minutes.
- Deterministic verdicts inject [JEV PRE-ANALYSIS] and set temperature 0 with maxTokens 400; the LLM still writes the answer, Jev only classifies.
- Fail-open guarantee: network, timeout or malformed Jev responses let the session proceed with normal reasoning.
- Config via JEV_API_KEY, JEV_API_URL (default https://api.typesafe.ai/v1/systemone) and JEV_MODEL (default jev-latest).
- MIT TypeScript 5.6 plugin for Node 18+; the README reports 51/53 tests green with typecheck OK.
Quickstart
bash
opencode plugin add github:murilogteixeira/opencode-jev-pre-llmWatch out
Grounded in the truncated repository README and GitHub metadata reviewed on 2026-10-15; the README reports 51/53 tests green but publishes no benchmarks or pricing, and its example prompt classifications are explicitly described as examples, not guaranteed verdicts. The project had 0 stars at review time.
Questions about this entry
Short answers drawn from the linked project or page.
An OpenCode V2 plugin that classifies each user prompt with TypeSafe Jev before the LLM reasons. Deterministic prompts get a JEV PRE-ANALYSIS block and temperature 0 so the model answers directly; everything else reasons normally.
It sends the prompt to the SystemOne API and applies two configurable thresholds: JEV_THRESHOLD (0-1, default 0.8) and JEV_MIN_CONFIDENCE (0-2, default 1.0). Both must pass before the plugin injects the pre-analysis block.
Run opencode plugin add github:murilogteixeira/opencode-jev-pre-llm, or clone the repo and npm install. Then set JEV_API_KEY; JEV_API_URL defaults to https://api.typesafe.ai/v1/systemone and JEV_MODEL defaults to jev-latest.
The plugin fails open by design: any Jev failure, whether network, timeout or an invalid response, lets the session proceed with normal LLM reasoning. The README states it never breaks the session.
No. The README is explicit that Jev only classifies (noul, choice or score); responses still come from the LLM, and reasoning is only suppressed when Jev confirms it is unnecessary.
One classification per prompt: the first prompt of a turn triggers Jev, and later steps in the same turn reuse the cached verdict with a configurable TTL, defaulting to 10 minutes.
murilogteixeira built it under MIT; the repo was created on 2026-10-11 and had 0 stars at review time. The README reports 51 of 53 tests passing with typecheck OK, and Node 18 or newer is required.
More Jev resources like this
A 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 Claude Code mod that adds $.jev to the engine so other mods can ask TypeSafe Jev typed Choice, Score and Noul questions and get calibrated probabilities back. Answers are validated against their questions, keys are kept in the keychain, and calls time out at 4 seconds by default.
Jev Integrations & Adapters#community#catalog-import#yibie
Community
gecm0/jev-modTypeScript
gecm0/jev-modTypeScriptAn opencode plugin that turns an agent's small decisions — is this command safe, which tool fits, is this PR ready — into typed Jev questions with calibrated confidences, labelling each answer auto or escalate so the agent knows when to ask for help.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
darwintechlab/openjevJavaScript
darwintechlab/openjevJavaScriptAn OpenCode plugin that asks Jev how much reasoning each step needs — low for a rename, high for a failing test — and adds the picked effort to the request, falling back to high when Jev is slow or unavailable. Archived and superseded by reasoning-router.
Jev Integrations & Adapters#community#catalog-import#logicrw
An OpenCode plugin that keeps the session on a cheap sticky parent model so the cache stays warm, and asks Jev whether a turn is hard before escalating through jev_escalate to temporary stronger subagents.
Jev Integrations & Adapters#community#catalog-import#cobanov
An OpenCode 2 plugin that sends every shell command — local or over FarHand's remote shell — to Jev first: six typed questions about global installs, host litter, harmful actions and privacy, and a permission prompt whenever something looks risky.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
cogflux/opencode-jev-guardTypeScript
cogflux/opencode-jev-guardTypeScriptFrom 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.