Jev glossary
PatternsContent moderation
Guardrail
Screening messages going into or out of an LLM app with hazard Nouls and severity Scores, then thresholding the probabilities.
- Category
- Patterns
- Also known as
- Content moderation
- Reading time
- 1 min
- Related terms
- 4
- Directory entries
- 129
- Docs
- docs.typesafe.ai
- Added
- 2026-09-24
Definition
The guardrails cookbook screens every message with one request: a Noul describes each possible hazard, a Score rates severity, and code thresholds the probabilities to pass, review, block, or route the message.
The RAG passage cookbook applies the same shape to retrieved context, using per-passage Nouls for relevance, evidence, contradiction, and injection before anything reaches the generator.
Tagspatternsguardrailssafety
Related terms
Definitions that connect to this one.
RAG passage classificationScore each retrieved passage with four Noul questions, then decide in code which passages reach the answering model.PatternsNoulThe yes/no primitive: one probability from 0 to 1, where the value is the answer and no separate confidence is returned.PrimitivesThresholdThe value your code compares against a score, a Noul probability, or a confidence to turn an answer into an action.Answers & ConfidenceAdversarial contentJev treats state as data, not as hostile input; injected instructions can move answers, so write precise criteria and test edge cases.Evaluation & Training
From the Jev directory
Directory entries that go deeper on this term.
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
A software factory that cleans AI-written code by combining Jev (System One), Gemini (System 2), and learned Semgrep rules. It verifies cuts with public and hidden tests, learns rules from verified decisions, and logs every decision.
Jev Demos & Experiments#madewithjev#agents#guardrails
CommunityShri2007/muscle-memory-aie-mlh
An autonomous research agent that searches grant sources daily, scores each lead's fit with Jev, and emails one ranked HTML digest with deadlines, amounts, and apply links. Python runs the loop; Composio supplies the search, page-fetch, and Gmail tools.
Jev Agents & MCP#agents#madewithjev#python
Swift package with two products: TypeSafeKit, a dependency-free HTTP client for the System One endpoint and model listing, and TypeSafeMiddleware, which adds a DecisionProvider protocol plus route, gate, rate, composite, and evaluate helpers for agent loops.
Jev Repos & SDKs#sdk#agents#routing
CommunityAgentiLoop/TypeSafeKit
Claude Code skill that reviews a finished video before publishing: Jev checks every frame against character reference sheets, Gemini watches with sound, GPT reviews still frames, and Claude returns a timestamped repair list.
Jev Agents & MCP#claude-code#agents#guardrails
CommunityArkya-AI/claude-video-review-skill
123 more matching entries in the full directory.
From the Jev AI model community
Posts from builders shipping with Jev right now.