Jev directory
CommunityJev Agents & MCP
Pi TypeSafe — npmjs.com
DevMortimer's Pi extension hands batched Choice, Score, and Noul judgments to Jev: a typesafe_evaluate agent tool, a terminal playground, and a typed API other extensions reuse, with daily request, token, and dollar caps.
- Reading time
- 2 min
- Published by
- Community
- Author
- DevMortimer
- Last verified
- Oct 2026
This is one of several Jev listings for the same project. The canonical entry is Pi TypeSafe — DevMortimer.
Used inJev for Software & DevTools
- Languages
- TypeScript
- Platforms
- CLI
- Capabilities
- Open sourceRequires API key
Highlights
- Requires Pi 0.85 or newer and Node.js 22.19 or newer; install with pi install npm:pi-typesafe.
- typesafe_evaluate is registered at startup but disabled until /typesafe enable, or PI_TYPESAFE_ENABLED=1 for headless runs.
- One batched call takes JSON state plus 1 to 32 questions and returns answers, model, token usage, and elapsed time.
- Daily caps bound requests, input tokens, and estimated spend; the environment can lower a cap but never raise it.
- The key is verified and saved to ~/.pi/agent/pi-typesafe/auth.json with owner-only permissions; TYPESAFE_API_KEY takes precedence.
Watch out
Independent project by DevMortimer, not affiliated with TypeSafe AI or the Pi authors, and the README says Jev access may be limited at the moment. Usage is billed to your TypeSafe account, and an enabled extension with no usable key looks exactly like a working one unless you check /typesafe status.
Questions about this entry
Short answers drawn from the linked project or page.
Run pi install npm:pi-typesafe, then /typesafe login with a key from console.typesafe.ai, /typesafe test to send one built-in sample request, and /typesafe enable to let the agent call the tool for the session.
It hands small structured judgments (classify, triage, compare, score) to Jev and returns calibrated numbers in one batched call, accepting JSON state plus 1 to 32 questions and returning answers, model, token usage, and elapsed time.
No. typesafe_evaluate is registered at startup but disabled until you run /typesafe enable and confirm the data notice. For automated or headless runs, set PI_TYPESAFE_ENABLED=1 explicitly.
After /typesafe login the key is verified against the API and saved to ~/.pi/agent/pi-typesafe/auth.json with owner-only permissions. For CI or scripts, TYPESAFE_API_KEY in the environment takes precedence over the stored key.
maxRequestsPerDay, maxInputTokensPerDay, and maxUsdPerDay bound a local day, survive restarts, and stop a request before it is submitted. The environment may lower an explicit cap but never raise it; a reached cap raises a budget error naming the cap, amount used, and day.
Whatever TypeSafe bills for the input tokens of your state and questions; output is free. At the listed rate of $42 per billion input tokens at the time of writing, a few hundred tokens cost well under a hundredth of a cent, and the per-session cap is 20 attempts.
Choice, Score, and Noul questions run in parallel and in isolation against the same state, so adding questions barely changes latency. evaluateAll chunks past the backend's limit of 32 questions (16 on SubQ) and fans them out.
More Jev resources like this
Amidabuddha's local stdio MCP server wraps TypeSafe's Jev in a single jev_decide tool: batched choice, score, and noul judgments with probabilities and token usage, leaving thresholds and actions to the caller.
Jev Agents & MCP#community#mcp#typescript
Community
Pi extension that registers MiniMax, StepFun, and Command Code as custom providers and adds a typesafe_decide tool answering typed Noul, Choice, and Score questions through TypeSafe Jev, with a Command Code fallback.
Jev Agents & MCP#community#typescript#pi
Community
An independent agent skill that walks a coding agent through designing a bounded Jev decision layer: contract, criteria, error behavior, fallback, and evaluation plan first, then scaffolding through TypeSafe or OpenRouter.
Jev Agents & MCP#community#catalog-import#logicrw
Community
grandamenium/jev-anythingPython
grandamenium/jev-anythingPythonFive typed tools — classify, check, score, rank, and ask — plus a skill, wired into Claude Code, Codex, Pi, and OpenCode in one setup command.
Jev Agents & MCP#community#typescript#mcp
Community
FrancoisChastel/jev-codeTypeScript
FrancoisChastel/jev-codeTypeScriptA DeepSeek Harness plugin that exposes Jev's noul, choice and score primitives as agent tools, plus configurable guardrails, retries, response caching, a circuit breaker and Web UI result cards.
Jev Agents & MCP#community#typescript#deepseek
Community
jackie-cqz/dsh-jev-pluginTypeScript
jackie-cqz/dsh-jev-pluginTypeScriptA Pi extension that reviews shell commands before execution in three tiers — auto-permitted, auto-blocked, and model-reviewed — with Jev, OpenRouter, or a pi model returning ALLOW or BLOCK for risky commands.
Jev Agents & MCP#community#typescript#pi
Community
From 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.