Source Available

Your coding agents lose context between sessions. Lerim helps them remember and learn continuously.

Continual learning layer for coding agents. Connect any agent, extract decisions and learnings automatically, and keep context fresh with background consolidation, deduplication, and forgetting.

Claude Code
Claude Code
Cursor
Copilot
OpenAI Codex
OpenAI Codex
Kiro
Kiro
Windsurf
Windsurf
Aider
Aider
VS Code
Cline
Cline
OpenCode
OpenCode
Zed
Amp
Amp

AI coding agents are stateless. Every session starts from zero. Decisions get re-debated. Patterns get re-discovered. Context gets lost.

Lerim helps agents remember what matters and keep learning across sessions.

Your agents learn.
Even while you sleep.

01

Agents work. Lerim watches.

Every coding session -- across Claude Code, Cursor, Codex, or any supported agent -- is automatically captured. No manual tagging, no workflow changes.

02

Decisions and learnings extracted.

LLM pipelines turn raw conversations into durable learning records -- architectural decisions, patterns discovered, pitfalls learned. Stored as plain markdown in your repo.

03

Every agent starts smarter.

Any agent can query shared context. Start a new session in a different tool and it can build on your project's history, conventions, and past decisions.

Everything you need for
continual agent learning

File-first learning store

Human-readable markdown in your repo. Git-friendly, no lock-in. git diff what your agents have learned.

Multi-agent support

Claude Code, OpenAI Codex, Cursor, OpenCode. One continual learning layer across all your tools.

Natural language chat

Ask questions about past decisions and patterns. Evidence-cited answers from your project learning history.

Background daemon

Auto-syncs new sessions as they appear. Background runs continuously refine context with consolidation, deduplication, and forgetting.

Consolidation and forgetting

Automatically merges duplicates, consolidates related learnings, and archives low-value or stale entries so context stays useful.

Web dashboard

Browse sessions, learning entries, and system health. Visual learning graph explorer.

Install once. Agents keep learning.

Recommended

Install as a skill

Add Lerim as a skill to any supported agent. It runs in the background automatically -- syncing sessions, extracting learnings, consolidating overlap, and forgetting low-value noise with zero manual effort.

$ npx skills add lerim-dev/lerim-cli
  Installed lerim skill.
  Works with: Claude Code, Cursor, Copilot, Codex...

Your agent now knows how to use Lerim. It can recall relevant context before a task and write back new learnings after it finishes -- across every session and every tool.

Or use the CLI

Full control from the terminal. Sync sessions, run consolidation and forgetting, and query your project's learning history directly.

$ pip install lerim
$ lerim sync
  Extracted: 8 decisions, 15 learnings
$ lerim maintain
  Merged: 3 duplicates. Consolidated: 2 related. Archived: 2 low-value.
$ lerim chat "Why SQLite over Postgres?"
  Decision [dec-041]: Chose SQLite for the index...

Plain markdown.
No lock-in.

Your agent learning records live as human-readable markdown files in your repo. Version-controlled, portable, and future-proof.

git diff what your agents learned.

.lerim/
memory/
  decisions/    # architectural choices
  learnings/    # patterns & pitfalls
  summaries/    # session records
  archived/     # retired entries
config.toml       # project settings