telos-mcp
Task planning and tracking for AI agents with project-based organization and Obsidian-compatible markdown storage.
README
telos-mcp
Task planning and tracking for AI agents. An MCP server with project-based organization and Obsidian-compatible markdown storage.
What it does
Gives any MCP-compatible AI agent persistent task management:
- Create, update, list, delete tasks with full status tracking
- Projects — organize tasks by project with metadata (tech stack, repo URL, sources)
- Dependencies — blockedBy/blocks relationships between tasks
- Bulk operations — create multiple tasks at once
- Task tree — hierarchical view of tasks with parent-child relationships
- Markdown files — each task is a folder with README.md and YAML frontmatter
- Obsidian integration — full compatibility with Obsidian vaults, Dataview, and wiki links
- Zero native addons — pure JavaScript, works on any Node version without rebuild
- Multi-agent — owner tracking so you know who's doing what
Quick start
npx telos-mcp
Stable install
npm install -g telos-mcp
Or from source:
git clone https://github.com/skye-flyhigh/telos-mcp.git
cd telos-mcp && npm install && npm run build
Then point your client to "command": "node", "args": ["/path/to/telos-mcp/dist/index.js"].
Configuration
| Variable | Default | Description |
|---|---|---|
TELOS_DIR |
~/.telos |
Directory where task files are stored |
Set TELOS_DIR to override the default storage location:
TELOS_DIR=/custom/path npx telos-mcp
Or in the MCP config:
{
"mcpServers": {
"telos": {
"command": "npx",
"args": ["telos-mcp"],
"env": { "TELOS_DIR": "/custom/path" }
}
}
}
Client Setup
Claude Code
Add to .claude.json:
{
"mcpServers": {
"telos": {
"type": "stdio",
"command": "npx",
"args": ["telos-mcp"],
"env": {}
}
}
}
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"telos": {
"command": "npx",
"args": ["telos-mcp"]
}
}
}
Tools
Task Management
| Tool | Description |
|---|---|
task_create |
Create a new task with subject, description, tags, dependencies, project_id |
task_update |
Update status, dependencies, project, or details of an existing task |
task_get |
Get full details of a task by ID |
task_list |
List all tasks (summary view), with optional filters |
task_delete |
Delete a task by ID (removes the folder) |
task_move |
Move a task to a different project (or to global tasks) |
task_create_bulk |
Create multiple tasks at once |
task_tree |
Get task hierarchy tree with all descendants |
Project Management
| Tool | Description |
|---|---|
telos_project_create |
Create a new project for organizing tasks |
telos_project_get |
Get full project information by key |
telos_project_list |
List all projects with optional filtering |
telos_project_update |
Update project metadata |
telos_project_archive |
Archive a project (moves to archive/year/) |
Status Workflow
pending → planning → in_progress → reviewing → completed
↓ ↓ ↓ ↓
blocked archived deleted
Use status: "deleted" in task_update to remove a task (same as task_delete).
Use status: "archived" to preserve task history without active tracking.
Dependencies
Tasks can block each other:
blockedBy: [1, 3]— this task can't start until tasks 1 and 3 complete- When a task is completed, it's automatically removed from dependents'
blockedBy - When a task is deleted, all dependency references are cleaned up
Storage Format
Directory Structure
~/.telos/
├── tasks/ # Global tasks (no project)
│ └── {id}-{slug}/
│ └── README.md
├── projects/
│ └── {key}/ # e.g., "mnemo", "black-cat"
│ ├── project.md # Project metadata
│ └── tasks/
│ └── {id}-{slug}/
│ └── README.md
└── archive/
└── {year}/
└── {key}/ # Archived projects preserved
├── project.md
└── tasks/
Task Format
Each task is a folder with a README.md containing YAML frontmatter:
---
id: 1
project_id: mnemo
subject: Fix authentication bug
status: in_progress
created: 2026-03-06T10:00:00.000Z
updated: 2026-03-06T12:30:00.000Z
owner: claude
activeForm: Fixing authentication bug
blockedBy: [3]
blocks: [5, 7]
tags: [backend, auth]
depth: 0
sources:
- https://docs.example.com
- https://github.com/org/repo/issues/42
---
Detailed description with full markdown support.
Project Format
Each project is a project.md file with YAML frontmatter and markdown body:
---
key: mnemo
display_name: Mnemo MCP
tech_stack: [typescript, mcp]
repo_url: "https://github.com/skye-flyhigh/mnemo"
status: active
created: "2026-03-06T10:00:00.000Z"
updated: "2026-03-06T12:00:00.000Z"
archived_at: null
archived_reason: null
sources:
- https://modelcontextprotocol.io
---
Memory management for AI agents. Description goes here as markdown body.
Obsidian Integration
Telos is designed for Obsidian compatibility:
- Markdown files — Every task is a README.md, every project is a project.md
- YAML frontmatter — Metadata in standard YAML format
- Wiki links —
[[project-key]]links between tasks and projects - Dataview queries — Dynamic tables for task and project lists
- Graph view — Visual task ↔ project relationships
Open ~/.telos as an Obsidian vault for full visualization.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.