Jev glossary
Evaluation & TrainingPrompt injection
Adversarial content
Jev treats state as data, not as hostile input; injected instructions can move answers, so write precise criteria and test edge cases.
- Category
- Evaluation & Training
- Also known as
- Prompt injection
- Reading time
- 1 min
- Related terms
- 4
- Directory entries
- 15
- Docs
- docs.typesafe.ai
- Added
- 2026-09-24
Definition
The jaggedness page is candid: content written to steer the model (an injected instruction, a misleading framing, text arguing for its own classification) can move the answer, and TypeSafe expects to improve on this.
The guardrails cookbook builds a filter around this with Noul questions, but the docs also warn that a passage filter is not a security boundary, because content below threshold can still reach a downstream prompt.
Tagslimitationssafety
Related terms
Definitions that connect to this one.
JaggednessThe documented set of jev-1.13 failure modes (literal reading, counting, dates, indirection, context rot), each with a guardrail.Evaluation & TrainingGuardrailScreening messages going into or out of an LLM app with hazard Nouls and severity Scores, then thresholding the probabilities.PatternsRAG passage classificationScore each retrieved passage with four Noul questions, then decide in code which passages reach the answering model.PatternsInstructionsThe field that states what a question should judge; accepts a string, an object, or an array for extra structure.Primitives
From the Jev directory
Directory entries that go deeper on this term.
A maintained list of jev-1.13's known failure modes, literal reading, unreliable counting, dates as text, indirection, context rot, and contradictory criteria, each with a guardrail.
Jev Practices & PatternsDocs#official#docs#evaluation
Official
Screen every message going into and out of an LLM app: a Noul describes possible hazards, a Score rates severity, and code thresholds the probabilities.
Jev CookbooksDocs#official#cookbook#guardrails
Official
Score retrieved passages with four Noul questions each, then route in code: prompt injections dropped, false premises kept as conflicts, and evidence assembled separately for the generator.
Jev CookbooksDocs#official#cookbook#search
Official
An opencode plugin (also for pi and Claude Code) that puts every bash command in front of a System One decision model before it runs: a small model scores destructiveness, secret leakage, and your repo's GUARDRAILS.md policy and blocks bad calls.
Jev Integrations & Adapters#community#github#opencode
Community
berget-ai/guardrails-mdTypeScript
berget-ai/guardrails-mdTypeScriptAn auto-mode safety monitor for the Pi coding agent: every tool call Pi is about to run gets one System One classifier question per rule, then allow, block or ask is decided in code. Built with DSPy against TypeSafe's Jev and exported to a Pi extension calling lunaroute/djev by default.
Jev Agents & MCP#community#github#pi
An MIT-licensed TypeScript bridge that brings System-1 decision models into Prime Agent v0.9.8+ as a decision layer with safety governance and a host bridge. The README advertises 185 passing tests at v0.1.0 and POSIX-only support.
Jev Integrations & Adapters#community#github#typescript
9 more matching entries in the full directory.
From the Jev AI model community
Posts from builders shipping with Jev right now.