---
title: "ledger-memory"
description: "Ledger-backed persistent memory for coding agents: append-only facts with retraction vaccines, scars, and a generated MEMORY.md projection"
canonical: https://agentpluginsdirectory.com/plugins/ledger-memory
last-updated: 2026-10-02
---

# ledger-memory
Ledger-backed persistent memory for coding agents: append-only facts with retraction vaccines, scars, and a generated MEMORY.md projection
- Slug: ledger-memory
- Publisher: Jesse Vincent
- Repository: https://github.com/prime-radiant-inc/ledger-memory
- Manifest: plugin.json
- Version: 0.1.4
- License: MIT
- Category (editorial): agent-tooling
- Skills: 1 (ledger-memory)
- MCP servers: 0
- Stars: 12
- Repository created: 2026-08-16
- Repository last pushed: 2026-08-20
- Publisher type: Organization
- Listing: https://agentpluginsdirectory.com/plugins/ledger-memory
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What ledger-memory does, in the publisher's words

ledger-memory is a Claude Code plugin that gives an agent persistent, append-only memory backed by ledger, replacing the old per-project markdown-file memory system. Facts are ledger keys with a status field (current, retracted, archived), so retracting a wrong fact leaves a permanent vaccine instead of silently deleting it, and a generated MEMORY.md projection, never the store itself, is what actually loads into an agent's context. See the design spec for the full rationale and the spike eval (8/8 agents, zero tool errors) that shaped the final write path.

ledger v0.1.0+ on PATH:

or

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/prime-radiant-inc/ledger-memory/HEAD/README.md

## Skills

- ledger-memory: Use when saving, correcting, or curating persistent memory, recording facts for future sessions, retracting wrong memories, end-of-session housekeeping, or deciding what deserves remembering.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.