vellum-assistant
Personal AI assistant with eight memory types and multi-channel support; per the list, it has an optional Jev provider.
Overview
Vellum Assistant is a personal AI assistant from Vellum with eight types of memory (episodic, semantic, procedural, emotional, prospective, behavioral, narrative and shared), self-written personality files, hourly proactive check-ins, strict actor identity and sandboxed tool calls, and channels including macOS, Windows, iOS, web, voice, email, Telegram, Slack and Twilio.
According to the list summary, it includes an optional Jev provider that sends conversation state and explicit questions to TypeSafe. The fetched README does not describe that integration, so where Jev is used in the assistant and what questions it asks are not documented here. It is MIT licensed. The assistant also checks its notes hourly for unfinished or due items and notifies you on the right channel without interrupting an active conversation.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Other | Explicit questions about conversation state sent to the optional Jev provider(inferred) | — |
How it works
- The assistant maintains conversation state and memory.
- When configured, it sends state and explicit questions to Jev.
- Typed answers inform the assistant's behavior.
Strengths
- Jev is optional within a full assistant product.
Caveats
- The README does not document the Jev integration.
- Jev's role appears minor relative to the product.
At a glance
- Type
- Project
- Status
- production
- Runtime
- desktop, mobile and web assistant
- Has code
- yes
- License
- MIT
Use cases and integrations
How curated lists describe it
An optional Jev provider in Vellum Assistant sends conversation state and explicit questions to TypeSafe.
logicrw/awesome-jev-projects
Frequently asked questions
How does it use Jev?
Per the list, via an optional provider that sends conversation state and questions to TypeSafe; the README does not detail it.
Is Jev required?
No, it is optional.
What is Vellum Assistant?
A personal AI assistant with multiple memory types and channels.
Related projects
Android AI chat client and SillyTavern-compatible endpoint where Jev screens memories for relevance and serves as a judge tool.
LibraryproductionKotlinAI companion emotion system that reads emotions from the companion's own written thoughts with Jev probabilities via OpenRouter.
ProjectprototypeJavaScriptDeepSeek Harness memory-palace plugin with spaced recall; an optional Jev layer judges ambiguous merge, keep or defer cases.
PluginbetaTypeScript- question-kit★ 3
TypeScript standard library for System One models: ready-made Jev methods with confidence gates and a turn-by-turn service agent.
ProjectprototypeTypeScript Federated knowledge retrieval engine for agents that uses Jev to route queries to backends and re-rank evidence, with keyword fallback.
LibraryprototypePythonAstrBot group-chat plugin fork that adds a 50-70 ms System One check on whether the bot should chime in, after the original random gate.
DemobetaPython