LuizEduPP/rememb

---
The problem
Teams using agents at real velocity rarely fail because they lack generation. They fail because operating agents every day creates context debt:
- too much re-explaining project facts every session
- too little durable memory outside the chat window
- too little audit trail for why something changed
- too much noise when recalling the right context
Every team or solo developer operating agents professionally hits this wall:
Existing solutions often center on hosted memory layers, API keys, or opaque context pipelines.
What you actually need is to **resume the next session with the minimum correct context and a trail you can inspect**.
rememb is built around four memory problems:
- durable facts and decisions instead of session-only chat memory
- keyword search instead of rereading everything (agents judge relevance)
- non-destructive versioning instead of silent overwrites
- local-first audit trail for AI work, not opaque cloud logs
---
Install
---
Quick Start
With MCP (recommended)
Zero friction. No CLI commands. Native IDE integration.
**1. Add to your IDE's MCP config:**
Add repeatable --project /path/to/repo args for per-project stores alongside global ~/.rememb. Details: MCP_TOOLS.md.
**2. Restart your IDE.**
The agent can read stored context at session start, write durable memory when something changes, and search only when targeted recall is needed.
If you want rememb usage to stay consistent, add a rememb-specific instruction block in your IDE custom instructions or in the MCP client prompt that wraps the agent. The point is to make the agent route reads, writes, search, recovery, and maintenance through rememb instead of ad hoc prompt memory.
You can place that block in either of these places:
- IDE-level custom instructions
- the system prompt or instruction field of the MCP client that is calling rememb
In both cases, keep the scope explicit: these rules are about how the agent should use rememb, not about replacing the rest of your coding instructions.
For the exact copy-paste block, use the canonical rules section in MCP_TOOLS.md.
No extra storage setup, server config, or schema migration is required. In MCP mode, rememb auto-initializes ~/.rememb and optional --project stores on first use.
For the current public MCP tool list (18 tools) and descriptions, see MCP_TOOLS.md.
If you want multiple MCP clients on the same machine to reuse one already-running rememb process, start a persistent local SSE transport:
This keeps one MCP process alive, so repeated clients can connect through http://127.0.0.1:8765/sse and http://127.0.0.1:8765/messages/.
Do not put --transport sse inside a stdio MCP client config. stdio clients expect JSON-RPC on stdin/stdout; the SSE mode exposes an HTTP endpoint and must be started separately.
Local usage without MCP
---
How it works
A local store on disk. Your agent can read prior decisions, search by keywords and tokens, update entries without losing history, and restore previous versions without depending on a cloud memory service. Copy ~/.rememb/ anywhere to move the store.
These map to rememb_write, rememb_edit, and rememb_delete. For the full MCP surface, see MCP_TOOLS.md.
Search uses **keyword and token matching** over entry content and tags. rememb returns full matches; the agent applies semantic relevance judgment. No API keys, no cloud, no embedding model download at runtime.