Memoir

io.github.bighippomanPublished by GitHub account bighippomanv1.0.3Updated Oct 24, 2026

Structured session journals for AI agents. Persistent memory across sessions.

VerifiedSTDIODesktop onlyKnowledge & MemoryDeveloper Tools

Overview

AI-generated overview

Memoir is a local MCP server that keeps structured session journals for AI agents, so a new session can pick up where the last one stopped.

What you can do
  • Record attempts and their outcomes, blockers, blocker resolutions, and design decisions in the current session
  • Close the current session with an optional summary
  • Read a structured handoff of the last session: what was attempted, what is blocked, what was decided
  • Query past sessions and list unresolved or resolved blockers across sessions
  • Scope logs automatically by git root and prune old sessions with rolling retention
When to use it
Useful for long-running coding projects where each new agent session otherwise starts from scratch and repeats dead ends, failed approaches or forgotten decisions. It suits developers who want a lightweight, local record of what was tried and why.
Requirements
Runs locally over stdio, typically started with npx, so Node.js is needed. No API keys or external services; data is stored in a single SQLite file under the user's home directory. Optional environment variables MEMOIR_MAX_CONTENT, MEMOIR_MAX_OUTCOME, MEMOIR_MAX_ENTRIES and MEMOIR_MAX_SESSIONS tune size limits and retention.
Before you install
The assistant writes journal entries to a local database, and old sessions are pruned automatically once the rolling limit is reached, so earlier history can be lost. Entries may capture project details, so review what gets logged before sharing the database or its contents.

Installation

In SourceWeft

  1. Open Memoir in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

memoir-mcp

Structured session journals for AI agents. Persistent memory across sessions -- no more repeating dead ends.

When a session ends, all reasoning is lost -- what was tried, what failed, what's blocked. The next session starts from scratch and repeats the same mistakes. memoir logs it all and hands it off so the next session picks up where the last one left off.

Works with any MCP client: Claude Code, Cursor, Codex, Windsurf, and more.

[memoir-mcp MCP server]

Install

Claude Code

bash
claude mcp add memoir -s user -- npx -y memoir-mcp

Other MCP clients

bash
npx -y memoir-mcp

How it works

  • Automatic project detection -- identifies the project by its git root, so logs stay scoped without any configuration.
  • Implicit sessions -- a session is created automatically on first log. No setup step.
  • Rolling retention -- keeps the last 20 sessions per project (configurable). Old sessions are pruned automatically.

Tools

Write

ToolDescription
log_attemptRecord something that was tried and its outcome.
log_blockerFlag something that's stuck and why.
resolve_blockerMark a blocker as resolved with what fixed it.
log_decisionRecord a design or architecture choice and its rationale.
end_sessionClose the current session with an optional summary.

Read

ToolDescription
get_handoffStructured summary of the last session -- what was attempted, what's blocked, what was decided.
get_historyQuery past sessions (default: last 3, max 20).
get_blockersList unresolved (or resolved) blockers across all sessions.

Storage

Single SQLite file at ~/.memoir/memoir.db. No API keys, no external services.

Configuration

All limits are configurable via environment variables:

VariableDefaultDescription
MEMOIR_MAX_CONTENT500Max characters for content fields
MEMOIR_MAX_OUTCOME300Max characters for outcome/resolution fields
MEMOIR_MAX_ENTRIES50Max entries per session
MEMOIR_MAX_SESSIONS20Max sessions per project (rolling)

Example with custom limits:

bash
claude mcp add memoir -s user -e MEMOIR_MAX_CONTENT=1000 -e MEMOIR_MAX_ENTRIES=100 -- npx -y memoir-mcp

Handoff output uses a compact format to keep context window usage low.

License

MIT

Source: README.md at commit 7c31ada

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.3LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts