jev-chat-gate
Framework-independent participation gate letting group-chat agents answer untagged questions only when Jev finds a useful reason to join.
Overview
Jev Chat Gate adds selective, untagged participation to AI agents in group chats. Mention-only agents wait to be tagged, so open questions to the group or chances to add useful information pass by. Before starting an unsolicited answer, the gate evaluates the recent conversation and the agent's stated capabilities: is there an unanswered request or something concrete to contribute, has someone already answered, and would joining intrude on a conversation between people.
Direct tags and replies keep the usual answering path without relevance scoring, while respecting participation controls. Observed dialogue context can support a follow-up without a new tag, and the gate can reject a proactive contribution as redundant or intrusive. The main model still writes the answer and can choose silence; the README stresses that usefulness is a judgment, not a guarantee. It is framework-independent, distributed through GitHub release tags rather than npm, independent of TypeSafe, and MIT licensed.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Yes / no | Is there an unanswered request or concrete contribution the agent can make? | yes · no |
| Yes / no | Has someone already answered, or would joining intrude on a conversation between people? | yes · no |
How it works
- New group message arrives without a tag
- Gate evaluates recent conversation and agent capabilities
- Admit or reject an unsolicited answer
- If admitted, the main model writes (or declines)
Strengths
- Agents can help without constant tagging
- Checks for redundancy and intrusion
- Direct requests unaffected
Caveats
- Usefulness judgments can be wrong
- Not on npm; installed from GitHub tags
At a glance
- Type
- Project
- Status
- beta
- Language
- TypeScript
- Runtime
- library
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
A framework-independent participation gate for AI agents in group chats, powered by typed Jev decisions.
logicrw/awesome-jev-projects
Frequently asked questions
What does Jev decide?
Whether an agent should join an untagged group conversation.
Does it change tagged replies?
No, direct requests keep their normal path.
Who writes the answer?
The agent's main model, which can still choose silence.
Related projects
AutoGPT agent platform, listed as adding Jev decision blocks for typed routing, filtering, scoring and confidence-gated next actions.
ProjectproductionPythonDecision middleware for AI agents that uses local Laya to pick tools and judge results, falling back to an LLM when unsure.
ProjectbetaPythonTerminal Q&A app where Jev sets response policy and reviews drafts while OpenAI only writes, orchestrated with LangGraph.
AppprototypePythonSimulated espresso bar where an XState machine defines what is possible and Jev chooses the barista's next event.
DemoprototypeTypeScript- jev-chat-windows★ 802
Windows WeChat reply assistant: local offline OCR reads the chat, Jev judges intent and tension and ranks three drafted replies to fill in.
ProjectproductionPython AstrBot plugin where Jev decides whether the bot should join a group conversation; the main model still writes the reply.
ProjectproductionPython