AccInt
@maxbaluev
AccInt is a local Work Model and MCP server for persistent scored agent memory, retrieval, commitments, browser/runtime observations, and outcome-based learning across coding agents.
概览
What is AccInt?
AccInt is a local work model for coding agents that integrates scored memory, recursive retrieval, commitments, runtime/browser observations, and outcome feedback in one substrate. It exposes an MCP interface so agents can retrieve prior work, record decisions, and close the loop when reality validates or rejects an action.
How to use AccInt?
—
Key features of AccInt
- Scored memory for prioritizing agent experiences
- Recursive retrieval of relevant prior work
- Commitments to track intended actions
- Runtime and browser observations
- Outcome feedback to validate or reject actions
- MCP interface for integration with coding agents
Use cases of AccInt
—
FAQ from AccInt
What does AccInt do?
AccInt is a local work model for coding agents that provides scored memory, recursive retrieval, commitments, runtime/browser observations, and outcome feedback via an MCP interface.
Where can I find the repository and website?
Repository: https://github.com/maxbaluev/accreted-intelligence ; Website: https://accint.xyz
