laiso

laiso / cman

Public

Agentic Memory for Claude Code

claude-code
18
0
100% credibility
Found Mar 26, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

cman is a plugin for Claude Code that lets users search and get summaries of their past AI sessions, plans, and memory files using natural language or dedicated skills.

How It Works

1
🔍 Discover cman

You hear about cman, a handy tool that helps your AI coding assistant remember all your past work without any extra setup.

2
🛒 Add the helper

In your chat with the AI assistant, you simply tell it to grab cman from the marketplace and turn it on.

3
✅ Everything ready

Now your assistant can automatically look through your old chats, plans, and notes whenever you need.

4
💭 Ask naturally

Just type a casual question like 'What did I do yesterday?' and watch it pull together your recent activities.

5
📋 See your recap

Get a clear, organized summary of finished tasks, ongoing work, and handy ways to jump back in.

6
📊 Check your status

Ask for a quick overview of active plans, recent sessions, and memory notes across your projects.

🚀 Stay on track

You effortlessly recall past work, resume sessions, and feel super organized with your AI sidekick.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is cman?

cman brings agentic memory to Claude Code, letting you search and summarize your existing session logs, plans, and memory files with natural queries like "What did I do yesterday?" or "Where was that auth work?". Built in Python as a plugin, it installs via the marketplace—no databases, APIs, or extra storage required—and delivers recaps with resume commands via skills like /cman:cm-search and /cman:cm-status. It solves the chaos of scattered Claude sessions across projects, turning them into an instant agentic memory bank.

Why is it gaining traction?

Unlike heavy agentic memory systems needing setup, cman agentically indexes your Claude files on-demand, outputting clean lists of plans, recent sessions, and keyword matches with cd && claude --resume shortcuts. In agentic memory reddit threads and surveys, devs praise its zero-overhead fit for agentic github copilot flows, outperforming manual grep or basic agentic memory benchmarks. The hook: Claude handles the summarization, making past work feel like live context.

Who should use this?

Claude Code users buried in dozens of sessions per project, like backend devs debugging infra upgrades or full-stack teams tracking web-app features across weeks. Perfect for solo hackers resuming stalled tasks or leads reviewing team progress without digging directories. Skip if you're not on Claude CLI daily.

Verdict

Try the cman repo if Claude Code is your workflow core—its agentic memory for llm agents shines in previews, but 18 stars and 1.0% credibility score signal early maturity with room for polish. Solid docs make it low-risk to test; watch for the cman repository github latest version.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.