Agent Memory

MCP server giving coding agents persistent, queryable architectural memory of a codebase.

Atom is a full-system code agent with root-level access, built for isolated servers.

Centralized memory for AI agents over MCP or plain CLI: remember/recall notes and work sessions across Claude Code, Codex, opencode, and more.

Local-first external brain CLI for coding agents — SQLite-backed context, graph-aware recall, and governed memory for Claude Code, Codex, and MCP clients.

A cognitive layer for Cursor's AI agent: lossless token savings via a content-addressed evidence store, plus verifiable semantic memory, project world model and skill runbooks. Fully local, zero telemetry. | 给 Cursor agent 的认知层：内容寻址证据库无损省 token；可验证语义记忆、项目

Engram — a self-hosted team knowledge wiki with semantic search, served as an MCP server. Propose-and-approve knowledge flow (human gate), shared data with the Engram desktop app.

Graph memory for AI agents — context engine for OpenClaw + MCP server for Claude Code/Claude.ai

Shared team memory for AI coding agents — MCP server and CLI for persistent, queryable knowledge across sessions and teammates

Portable agent working state MCP server — tasks, plans, notepads, memory, and rules for any AI coding agent

Grounded, self-checking document memory as a Model Context Protocol server: store text, ask questions, get cited answers or an honest "not enough evidence".

MCP server for Lithium. Give AI agents tools to navigate, store, and retrieve structured data.

MCP server for persistent, cross-agent working memory. BYO brain directory.

Agent metacognitive monitoring and calibration — knowledge boundary detection, Ebbinghaus forgetting alerts, consolidation triggers, self-reflection, and 3-stream transformer confidence calibration (ECE + Brier)

Personal Knowledge Brain for developers — MCP server with typed neurons, hybrid search (FTS5 + vector), knowledge graph, vault encryption, time-travel history, and spaced repetition. 100% local, SQLite-based.

Governed local-first memory for coding agents

Continuous learning for Kiro agent sessions on AWS. Passive tool-use capture, LLM extraction, and synchronous context injection — inspired by claude-mem, aligned with Bedrock AgentCore Memory.

Local-first biological memory for coding agents: recall, consolidation, and pre-action Guard checks.

Zettelkasten-based agent memory system with bidirectional links

Local-first MCP server that gives AI coding agents a living, browser-viewable project wiki and a project-local memory store, so the agent stays oriented and the human stays in control.

El Primor — AI coding agent for OpenCode. 9-phase methodology, 7 expert skills, persistent memory (TurboVec). Zero-friction install.