wrm3

wrm3 / galdr

Public

AI-powered development framework with task management, 41 agents, 83 skills, and MCP tools for Cursor, Claude Code, Gemini, Codex & OpenCode. File-based memory that survives across sessions.

agent-framework agent-swarm ai-agents ai-framework ai-memory
13
0
100% credibility
Found Mar 31, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

galdr is a file-based framework that equips AI coding agents with persistent memory, structured task tracking, architectural rules, and specialized skills across multiple AI IDEs.

How It Works

1
🔍 Discover galdr

You hear about galdr, a helper that makes AI coding buddies remember your project details, tasks, and rules so they don't forget between chats.

2
Choose your setup
👥
Join team setup

Connect to your team's shared helper service.

🛠️
Set up yourself

Download and start your own helper with a quick launch.

3
💻 Open in your coding app

Load your project in Cursor, Claude Code, or your favorite AI coding tool.

4
🧠 Agents remember everything

Your AI buddies instantly know your tasks, past work, rules to follow, and project goals—no repeating yourself.

5
📝 Assign tasks and watch magic

Create checklists or say 'work on this,' and agents plan, code, review, and track progress automatically.

🎉 Projects flow smoothly

AI agents stay smart across chats and tools, finishing work faster while always respecting your rules and preferences.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 13 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 galdr?

Galdr is a Python framework that turns forgetful AI coding agents into reliable teammates by adding file-based persistent memory, task tracking, and 83 specialized skills across IDEs like Cursor, Claude Code, Gemini, and Codex. It stores project context, tasks, and constraints in a .galdr/ folder that survives sessions and machines, with Docker MCP tools for RAG search, video analysis, and vault knowledge. Developers invoke it via commands like @g-task-new or @g-review for structured AI-powered development workflows.

Why is it gaining traction?

Unlike one-off chats, galdr enforces architectural rules agents must acknowledge and syncs task status across IDEs, eliminating repetition. Its 41 agents handle everything from sprint planning to platform doc crawling, with WebSocket dashboards for team visibility. As an AI-powered development platform, it bridges Cursor workshops to production via shared memory and MCP integration.

Who should use this?

AI enthusiasts building agentic apps in Cursor or Claude Code, needing task backlogs that persist across reboots. Teams running AI-powered development services with video analyzers or galdr tables for Valheim-like project graphs. Solo devs tired of re-explaining constraints in multi-IDE setups.

Verdict

At 13 stars and 1.0% credibility, galdr feels experimental—docs are solid but lacks broad testing. Worth a spin for AI-powered projects on GitHub if you're into Cursor agents; skip for mission-critical work until it matures.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.