Abscissa
A Python stdio MCP server for Linear's GraphQL API, providing 35 tools to manage issues, projects, cycles, dependencies, comments, labels, workflow states, teams, and users with cursor-paginated results and destructive action confirmation.
README
<p align="center"> <img src="https://raw.githubusercontent.com/KazKozDev/abscissa/main/docs/assets/banner.png" alt="Abscissa" width="350"> </p>
<!-- mcp-name: io.github.KazKozDev/abscissa -->
Abscissa is a Python stdio MCP server for Linear. It turns Linear's GraphQL API into 42 tools for an MCP client: issues, projects, cycles, dependencies, comments, labels, workflow states, teams, and users.
The design priority is deliberate control over project data. Read operations
return cursor-paginated results; get_user_issues() resolves the authenticated
Linear user; archive and delete tools require an explicit confirm=true before
they make a destructive API call. The server stores no credentials and reads
LINEAR_API_KEY from its process environment.
Abscissa uses the stdio transport from the Model Context Protocol. Any client that supports stdio MCP can launch it as a local tool process.
Run it
You need Python 3.10+ and a Linear personal API key.
git clone https://github.com/KazKozDev/abscissa.git
cd abscissa
python3 -m venv .venv
.venv/bin/pip install -e '.[dev]'
export LINEAR_API_KEY='lin_api_…'
.venv/bin/abscissa
You can also install Abscissa directly from PyPI:
pip install abscissa
The final command starts the MCP server on standard input and output. Register
.venv/bin/abscissa as a stdio command in your MCP client, and pass
LINEAR_API_KEY through that client's environment or secret manager.
The key remains outside the repository. The .gitignore excludes common local
environment files, including .env and .venv.
Example prompts
Once Abscissa is registered in your MCP client, try prompts like:
Show my open Linear issues grouped by project.
Create an issue in ENG for fixing onboarding copy.
List issues blocked by GEN-32.
Tools
| Area | Tools |
|---|---|
| Identity | get_current_user |
| Issues | create_issue, update_issue, search_issues, get_user_issues, get_issue, assign_issue, add_comment, set_issue_estimate |
| Issue lifecycle | archive_issue, delete_issue |
| Dependencies | list_issue_dependencies, add_issue_dependency, remove_issue_dependency |
| Labels and workflow | list_workflow_states, list_issue_labels, create_label, add_issue_label, remove_issue_label |
| Teams and people | list_teams, get_team, list_users |
| Projects | create_project, get_project, list_projects, list_project_issues, set_issue_project, remove_issue_from_project, update_project, update_project_details, archive_project |
| Project labels | list_project_labels, create_project_label, add_project_label, remove_project_label |
| Project updates | create_project_update, list_project_updates |
| Cycles | create_cycle, update_cycle, archive_cycle, list_cycles, list_cycle_issues |
Search and list tools accept limit and cursor. Their responses keep the same
shape:
{
"items": [],
"next_cursor": null
}
Pass a non-null next_cursor back as cursor to request the next page.
list_issue_dependencies returns separate blocks and blocked_by lists.
Each item includes its relation_id; use it to remove that dependency. Its
next_cursors object has separate cursors for each direction.
Destructive actions
archive_issue, archive_project, and delete_issue are marked with MCP's
destructiveHint. They refuse to call Linear until the client invokes them
with confirm=true.
Deletion requires explicit confirmation: set confirm=true
This prevents an accidental tool call from deleting or archiving data. It does not replace Linear's own access controls: the API key still determines which resources the server may read or change.
Verify the checkout
.venv/bin/ruff check .
.venv/bin/python -m pytest
The current checkout produces:
All checks passed!
.......... [100%]
10 passed
The tests cover pagination bounds and response shape, authenticated-user issue resolution, project membership, cycle queries, dependency direction, confirmation guards, cycle lifecycle operations, and MCP tool registration. GitHub Actions runs the same lint and test commands on Python 3.10, 3.11, 3.12, and 3.13.
Limitations
Abscissa is a local stdio server, not an HTTP service. It needs a running MCP-capable client and a valid Linear API key. The automated suite avoids mutating a real Linear workspace; create, update, archive, and delete behavior is therefore protected by unit tests rather than a live write test.
License
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.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
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.