Graph-tl/graph
Agent-native persistent task graph. MCP server for agent planning and execution across sessions.
Implements dependency-aware task planning with automatic unblocking, storing architectural decisions and naming conventions as reusable knowledge that surfaces across sessions. Operates as an MCP server with a local SQLite backend, exposing 22 tools for planning, execution tracking, and structured handoffs—agents call `graph_onboard` to retrieve full project context, claim tasks via `graph_next`, and record evidence on completion to unlock dependent work.
Available on npm.
Stars
6
Forks
2
Language
TypeScript
License
MIT
Category
Last pushed
Mar 04, 2026
Commits (30d)
0
Dependencies
3
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/mcp/Graph-tl/graph"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
cjo4m06/mcp-shrimp-task-manager
Shrimp Task Manager is a task tool built for AI Agents, emphasizing chain-of-thought,...
stanislavlysenko0912/todoist-mcp-server
Full implementation of Todoist Rest API & support Todoist Sync API for MCP server
democratize-technology/vikunja-mcp
Model Context Protocol server for Vikunja task management. Enables AI assistants to interact...
themotionmachine/OmniFocus-MCP
Let LLMs interface with your tasks and projects through the Model Context Protocol. Add,...
jpicklyk/task-orchestrator
A light touch MCP task orchestration server for AI agents. Persistent work tracking and context...