LangChain
LangChain's Python framework, which includes an optional TypeSafeClassifier runnable plus model-routing and risky-tool middleware using Jev.
Overview
LangChain is the widely used framework for building agents and LLM-powered applications, chaining interoperable components and third-party integrations. Its ecosystem includes Deep Agents, LangGraph for controllable agent workflows, many provider integrations and LangSmith for evaluation and observability.
Per the listings, LangChain's Python package includes an optional Jev integration: a TypeSafeClassifier runnable that sends batched typed questions to Jev, plus middleware for model routing and for gating risky tool calls. The README excerpt is LangChain's general overview and does not describe this integration, so its exact API is not confirmed here. Its inclusion signals that a major agent framework treats typed decision models as a component for fixed harness roles such as routing and approval rather than open-ended generation.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which model should handle this request? (model-routing middleware)(inferred) | — |
| Yes / no | Is this tool call risky enough to block or confirm? (risky-tool middleware)(inferred) | — |
| Choice | User-defined typed classification via TypeSafeClassifier | — |
How it works
- Add the TypeSafeClassifier runnable or middleware to a LangChain app
- Runnable sends batched typed questions to Jev
- Routing middleware selects a model; risky-tool middleware gates calls
- Chain continues with the typed result
Strengths
- Integration inside a major agent framework
- Batched typed questions
- Reusable middleware for routing and tool gating
Caveats
- Integration details not in the main README
- Optional component; requires a TypeSafe key
- LangChain itself is a large general framework
Getting started
uv add langchain; see LangChain's integration docs for the TypeSafe classifier.At a glance
- Type
- Plugin
- Status
- production
- Language
- Python
- Runtime
- LangChain framework
- Has code
- yes
- Has tests
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
An optional Jev classifier integration for Python LangChain workflows.
logicrw/awesome-jev-projectsPython `TypeSafeClassifier` Runnable with batched typed questions plus model-routing and risky-tool middleware.
cobanov/awesome-jev
Frequently asked questions
How does LangChain use Jev?
Through an optional TypeSafeClassifier runnable and middleware for model routing and risky tool calls, per the listings.
Is it required?
No, it is an optional integration.
Where is it documented?
In LangChain's integration documentation rather than the main README.
Related projects
LangChain.js, which includes an optional TypeSafeClassifier runnable and middleware for bounded routing and tool-call checks with Jev.
PluginproductionTypeScriptLangChain blog guide on wiring Jev into an agent harness as a dangerous-action classifier and model router.
ArticlediscussionComposio agent SDK with an optional TypeSafe provider that uses Jev to shortlist and pick tools and closed-set arguments.
ProjectproductionTypeScriptClaude Code mod that adds $.jev to the engine interface so other plugins and hooks can ask Jev typed questions through six providers.
ServicebetaTypeScriptYielded Agent: durable Effect agent threads in TypeScript, listed for a TypeSafe decision provider for typed question sets.
ProjectbetaTypeScriptVercel's filesystem-first agent framework, whose experimental evaluate and autoModel paths default to Jev via AI Gateway.
ServicebetaTypeScript