@duytnb79/asana-mcp
A local MCP server for Asana that uses a personal access token to read and manage Asana projects, tasks, sections, and comments via the standard Asana REST API.
README
@duytnb79/asana-mcp
A local MCP server for Asana that uses an Asana personal access token (PAT).
It runs over stdio and calls the standard Asana REST API at https://app.asana.com/api/1.0. It is separate from Asana's hosted MCP server at https://mcp.asana.com/v2/mcp, which requires a registered MCP app and OAuth.
Requirements
- Node.js 20+
- An Asana personal access token
- Access to the Asana workspaces, projects, and tasks you want to use
Create a PAT in the Asana developer console and treat it like a password. The server can only access data and perform actions allowed for the Asana user who owns the token.
Installation
Local clone
npm install
npm run build
cp .env.example .env
node dist/index.js
Published package
After this package is published, it can be run with:
npx -y @duytnb79/asana-mcp
Or installed globally:
npm install -g @duytnb79/asana-mcp
asana-mcp
Configuration
Create a .env file or provide environment variables through your MCP client:
ASANA_ACCESS_TOKEN="your_asana_personal_access_token"
ASANA_TIMEOUT_MS="10000"
ASANA_MAX_PAGE_SIZE="100"
Required:
ASANA_ACCESS_TOKEN
Optional:
ASANA_TIMEOUT_MS— request timeout in milliseconds; defaults to10000ASANA_MAX_PAGE_SIZE— maximum page size exposed by list/search tools; defaults to100and must be between1and100
The server automatically loads .env when running locally.
MCP client configuration
Local build
{
"mcpServers": {
"asana": {
"command": "node",
"args": [
"/Users/genkisystem/Desktop/asana-mcp-server/dist/index.js"
],
"env": {
"ASANA_ACCESS_TOKEN": "your_asana_personal_access_token"
}
}
}
}
Alternatively, run through npm from the project directory:
{
"mcpServers": {
"asana": {
"command": "npm",
"args": ["start"],
"cwd": "/Users/genkisystem/Desktop/asana-mcp-server",
"env": {
"ASANA_ACCESS_TOKEN": "your_asana_personal_access_token"
}
}
}
}
Published package
{
"mcpServers": {
"asana": {
"command": "npx",
"args": ["-y", "@duytnb79/asana-mcp"],
"env": {
"ASANA_ACCESS_TOKEN": "your_asana_personal_access_token"
}
}
}
}
Available tools
Read tools
list_projects- Lists projects in a workspace.
- Supports
archived,limit,offset, andopt_fields.
list_tasks- Lists tasks in a project in project priority order.
- Supports
completed_since,limit,offset, andopt_fields.
get_task- Gets one task by GID.
search_tasks- Searches a workspace by assignee, completion state, modified time, project, or text.
- Asana search is eventually consistent and may lag recent writes by 10–60 seconds.
- The search endpoint does not support normal Asana offset pagination and returns at most 100 items.
list_sections- Lists sections in a project.
- Supports
limit,offset, andopt_fields.
Write tools
create_task- Creates a task or subtask.
- Requires at least one of
workspace,projects, orparent.
update_task- Updates fields on an existing task.
add_comment- Adds a plain-text comment to a task.
Write tools make immediate changes in Asana. Configure your MCP client to require approval for these tools if you want a human confirmation step.
Pagination
Asana list endpoints return an opaque next_page.offset. The MCP response exposes it as meta.next_offset.
Pass that value back as offset to retrieve the next page. Only use offsets returned by Asana; they can expire when underlying data changes.
Input/output fields
Asana returns compact objects by default. Use opt_fields to request additional properties, for example:
{
"project_gid": "12345",
"limit": 50,
"opt_fields": [
"name",
"completed",
"assignee.name",
"due_on",
"permalink_url"
]
}
Keep opt_fields focused. Very broad or deeply nested responses are more expensive and may be rate-limited.
Rate limits and errors
- Asana returns HTTP
429when a token is rate-limited. - The server reports the
Retry-Aftervalue when Asana provides it, but does not automatically retry write requests because retries could duplicate creations or comments. - Authentication, permission, validation, not-found, timeout, and server errors are converted into readable MCP errors.
- The PAT is sent only in the
Authorization: Bearerheader and is never placed in request URLs.
Security
- Never commit
.envor a PAT. - Prefer a dedicated PAT with the minimum user permissions needed for this integration.
- Rotate the PAT if it is exposed.
- Tool calls run with the permissions of the token owner.
- This server intentionally exposes specific Asana operations rather than a generic HTTP passthrough tool.
Development
npm run dev
npm run typecheck
npm run build
npm start
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.
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.
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.
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.