Venkai¶

Venkai is a persistent context layer for AI agents. Agents write down facts, decisions, preferences, events, constraints and relationships; Venkai keeps them across sessions, models and workflows, and hands back a ranked short list when the next agent asks for context.

Beta — API version 0.1.0, live at https://api.venkai.fr. See Status & maturity.

What problem does it solve?¶

An agent session ends and its reasoning evaporates. The next run re-derives the same constraints, re-asks the same questions, or contradicts a decision made yesterday. The usual patch — paste the whole history back into the prompt — buys continuity by spending the context window on text the model does not need.

Venkai splits those two things apart:

	Without Venkai	With Venkai
Write	Append to a transcript	Store a typed memory with an importance and metadata
Read	Replay everything	Get a ranked short list for a specific task

The read side is the product. GET /api/context/{project}/relevant?query=…&limit=5 returns at most limit memories, each with a score and a justification.

How does it work?¶

flowchart LR
  A[Your agent] -->|store what it learned| V[Venkai]
  V --> P[(Persistent context)]
  A -->|ask: what matters for this task?| V
  P --> R[Relevance + context signals]
  R -->|ranked, bounded list| A

Your agent writes memories when it learns something, and asks for context when it starts a task. Venkai ranks what it holds against that task and returns a list you size yourself. More in How Venkai works.

How do I integrate it?¶

Three surfaces, one API behind all of them:

REST API — Authorization: Bearer vk_live_…, any language.
Python SDK — client.memory.remember(...) / client.memory.recall(...).
MCP server — tools for Claude Code and any MCP-compatible client.

Start with the Quickstart — five minutes from an API key to a ranked retrieval.

Why is it different?¶

Venkai is deliberately not several things it is often mistaken for:

It is not	Because
An LLM	Venkai does not generate text. It never calls a model to answer.
An agent framework	No planning loop, no tool routing. It sits beside your framework.
A vector database	It returns a decision about what belongs in a prompt, not just nearest neighbours.
A prompt cache	Memories are typed, editable, versioned and organization-scoped, not keyed on a prompt.

Status and maturity¶

Venkai is pre-1.0. This documentation labels every capability:

Label	Meaning
Stable	Implemented, covered by tests, running in production
Beta	Implemented and running, API may still change
Experimental	Implemented but not validated; do not build on it
Planned	Not implemented. Documented so you can tell it apart from the rest.

There are no published performance benchmarks for Venkai yet. See Evaluation for how we approach claims, and how to evaluate Venkai on your own workload.