spranab

spranab / saga-mcp

Public

A Jira-like project tracker MCP server for AI agents. SQLite-backed, 22 tools, full hierarchy, activity logging, and dashboard.

12
2
100% credibility
Found Feb 24, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

saga-mcp is a local file-based project tracker that gives AI assistants structured tools to manage hierarchies of goals, tasks, checklists, notes, comments, and progress across sessions.

How It Works

1
🕵️ Discover saga-mcp

You hear about a handy organizer that lets your AI assistant keep track of projects without forgetting anything between chats.

2
📁 Pick a save spot

Choose a folder in your project where all your task info will live in one simple file.

3
🔗 Link to your AI

Tell your AI helper where that file is, and connect the organizer so it can use it.

4
🚀 Launch and go

With one easy start, your AI now has a permanent memory for all your work – everything feels structured and reliable.

5
📋 Start your project

Ask your AI to set up a new project, like planning a website or app, and watch it create the big picture.

6
✅ Build out the details

As you chat, your AI adds goals, steps, checklists, notes, and even links decisions together automatically.

7
📊 See the big picture

Get a quick dashboard view anytime with progress stats, what's blocked, and a friendly summary of it all.

🎉 Projects stay alive

Your work is forever organized, your AI picks up right where it left off, and you feel in total control.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 12 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 saga-mcp?

Saga-mcp is a Jira-like project tracker built as an MCP server for AI agents, using TypeScript and a local SQLite-backed database per project. It replaces scattered markdown files with a full hierarchy of projects, epics, tasks, subtasks, notes, comments, and dependencies, plus activity logging and a dashboard tool for overviews. AI agents like those in Claude can call 30+ tools to create, update, search, and export work without losing context across sessions.

Why is it gaining traction?

Unlike cloud Jira or Microsoft Project alternatives, saga-mcp runs offline with zero setup—just point DB_PATH to a .tracker.db file and npx it. Developers hook into its MCP tools for agent-driven workflows, with auto-blocking dependencies, template substitution, batch ops, and natural language dashboard summaries standing out for agent reliability. The activity log and session diff tools make resuming work seamless, turning AI into a persistent project nanny.

Who should use this?

AI agent users building saga-like workflows in Claude Desktop or codebases, especially solo devs tracking epics across sessions. Frontend or fullstack teams wanting a local Jira-like tracker for tasks, blockers, and decisions without subscriptions. Those experimenting with MCP servers for agent distribution in project management.

Verdict

Try saga-mcp if you're deep into AI agents—its docs and tool annotations are solid for quick starts, despite 10 stars and 1.0% credibility signaling early maturity. Pair it with better tests for production, but it's a smart SQLite-backed bet for agent-powered tracking now.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.