dfrancislyondflabc-tech

Agentic Recall — Documentation skill for Claude Code

Documentation community

Long-term memory for agentic tasks.

How to install Agentic Recall

This entry records only its repository, not the path inside it, so there is no exact command to give. Open dfrancislyondflabc-tech/agentic-recall and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agentic Recall does

Long-term memory for agentic tasks. It tells you when it doesn't know. An MCP server for Claude Desktop and Claude Code: hybrid retrieval over your own markdown memories, with an explicit absence verdict instead of a least-bad match.

Alternatives in Documentation

Claude Code Ultimate Guide — by Florian BRUNIAUX - A tremendous feat of documentation, this guide covers Claude Code from beginner to power 2k ★
Apply Self Review — Apply the approved improvements from the latest memory-self-review PROPOSAL 297 ★
Project Architect — Documentation-first project planning agent skill 250 ★

README

agentic-recall

**Long-term memory for agentic tasks. It tells you when it doesn't know.**

An MCP server that gives Claude Desktop and Claude Code a persistent memory across sessions — and, unlike a plain vector store, an explicit answer when the corpus does not contain what you asked for.

![agentic-recall: a query that hits, and a query that does not](docs/demo.gif)

// ~/.claude.json  (Claude Code)  or  claude_desktop_config.json  (Claude Desktop)
"mcpServers": {
  "memory": {
    "command": "node",
    "args": ["/absolute/path/to/agentic-recall/index.js"],
    "env": { "MEMORY_DIR": "/absolute/path/to/your/memory/folder" }
  }
}

git clone https://github.com/dfrancislyondflabc-tech/agentic-recall.git
cd agentic-recall && npm install && npm test    # 89+ checks, no corpus of your own needed

Requires **Node 20 or 22**. Nothing else — the embedding model downloads on first index and then runs locally. Your memories never leave the machine.

Built by Daniel Francis-Lyon — questions, bug reports and criticism all welcome, either as an issue on this repo or at **danfrancislyon@gmail.com**.

The part most memory tools skip

Retrieval always returns *something*. Ranked by similarity, the least-bad match comes back looking like an answer, and an agent acting on it cannot tell the difference between "here is what you wrote" and "here is the closest thing to it".

This server separates those two cases and says which one you got:

// asked about something that isn't in the corpus
{ "results": [],
  "confidence": "low",
  "bestWeak": [{ "name": "wheel-truing", "score": 0.31 }],
  "absenceNote": "no strong match — 'afternoon' appears nowhere in this corpus" }

Nothing was invented, nothing was silently ranked into first place, and the nearest thing is offered as a candidate rather than as an answer. Everything below is about how that verdict is reached and how it is measured.

Two-tier hybrid retrieval over Claude's persistent