Abscissa
Abscissa is a Python MCP server that exposes Linear's GraphQL API as 42 tools, enabling full project management from any MCP-compatible client.
Identity & Users
Get the authenticated Linear user and list all organization users.
Issues
Create, update, retrieve, and search issues (with filters like state, priority, assignee, team).
Get issues assigned to yourself or a specific user; assign issues to users.
Add markdown comments, set story point/hour estimates, and add/remove labels.
Archive or permanently delete issues (requires
confirm=true).
Issue Dependencies
List, add, and remove dependencies between issues (blocks/blocked by).
Labels & Workflow States
List team workflow states; list, create, attach, and remove issue labels.
Projects
Create, get, list, update, and archive projects (archive requires
confirm=true).List project issues; add or remove issues from projects.
Post and list project status updates with health indicators.
Project Labels
List, create, attach, and remove reusable project labels.
Teams
List all teams and get details for a specific team by key or UUID.
Cycles (Sprints)
Create, update, archive, and list cycles; list issues within a specific cycle.
Pagination
Most list and search tools support cursor-based pagination via
limitandcursorparameters.
Safety
Destructive operations (archive/delete) require explicit
confirm=trueto prevent accidental data loss.
Provides tools for interacting with Linear's API, enabling AI agents to manage issues, projects, cycles, dependencies, comments, labels, workflow states, teams, and users programmatically through the Linear platform.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Abscissa show my open issues"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
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/abscissaYou can also install Abscissa directly from PyPI:
pip install abscissaThe 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.
Related MCP server: linear-mcp
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 |
|
Issues |
|
Issue lifecycle |
|
Dependencies |
|
Labels and workflow |
|
Teams and people |
|
Projects |
|
Project labels |
|
Project updates |
|
Cycles |
|
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=trueThis 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 pytestThe current checkout produces:
All checks passed!
.......... [100%]
10 passedThe 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
Available Tools
42 toolsadd_commentC
Add a markdown comment to a Linear issue.
Inputs: body (required), issue_id (required). Returns: result
add_issue_dependencyC
Record that blocker_issue_id blocks blocked_issue_id in Linear.
Inputs: blocked_issue_id (required), blocker_issue_id (required). Returns: result
add_issue_labelC
Add a label to an issue.
Inputs: issue_id (required), label_id (required). Returns: result
add_project_labelB
Attach an existing project label to a project.
Inputs: label_id (required), project_id (required). Returns: result
archive_cycleADestructive
Archive a cycle. Set confirm=true only after explicit user confirmation.
Inputs: cycle_id (required), confirm. Returns: result
archive_issueADestructive
Archive an issue. Set confirm=true only after explicit user confirmation.
Inputs: issue_id (required), confirm. Returns: result
archive_projectADestructive
Archive a project. Set confirm=true only after explicit user confirmation.
Inputs: project_id (required), confirm. Returns: result
assign_issueC
Assign a user to an issue.
Inputs: user_id (required), issue_id (required). Returns: result
create_cycleA
Create a Linear cycle for a team.
starts_at and ends_at must be ISO 8601 timestamps, for example 2026-07-13T09:00:00Z. team_id accepts a team UUID or key.
Inputs: name (required), ends_at (required), team_id (required), starts_at (required), +1 more. Returns: result
create_issueB
Create a new Linear issue.
team_id accepts a team UUID or key (for example, ART). Priority is 0=none, 1=urgent, 2=high, 3=normal, or 4=low.
Inputs: title (required), team_id (required), state, priority, +1 more. Returns: result
create_labelC
Create a label for a team.
Inputs: name (required), team_id (required), color. Returns: result
create_projectA
Create a project; team_ids may be comma-separated team keys or UUIDs.
Inputs: name (required), team_ids, description. Returns: result
create_project_labelB
Create a reusable project label. Use add_project_label to attach it to a project.
Inputs: name (required), color, description. Returns: result
create_project_updateC
Post a project update. Health may be onTrack, atRisk, or offTrack.
Inputs: body (required), project_id (required), health. Returns: result
delete_issueADestructive
Delete an issue permanently. Set confirm=true only after explicit user confirmation.
Inputs: issue_id (required), confirm. Returns: result
get_current_userA
Return the authenticated Linear user behind the configured API key.
Inputs: no named parameters. Returns: result
get_issueA
Get full issue details, including comments, labels, assignee, and dates.
Inputs: issue_id (required). Returns: result
get_projectC
Get a project by UUID.
Inputs: project_id (required). Returns: result
get_teamA
Get a team by key (for example, ART) or UUID.
Inputs: identifier (required). Returns: result
get_user_issuesB
Get a user's assigned issues; omit user_id to retrieve the authenticated user's issues.
Inputs: limit, cursor, user_id. Returns: result
list_cycle_issuesB
List issues in a cycle. Pass next_cursor as cursor for the next page.
Inputs: cycle_id (required), limit, cursor. Returns: result
list_cyclesA
List workspace cycles, optionally filtered by a team key or UUID.
Inputs: limit, cursor, team_id. Returns: result
list_issue_dependenciesA
List issues this issue blocks and issues that block it.
The returned relation IDs can be passed to remove_issue_dependency. Use the matching cursor from next_cursors to retrieve another page for either side.
Inputs: issue_id (required), limit, blocks_cursor, blocked_by_cursor. Returns: result
list_issue_labelsA
List labels for a team. Pass next_cursor as cursor for the next page.
Inputs: team_id (required), limit, cursor. Returns: result
list_project_issuesA
List a project's issues. Pass next_cursor as cursor for the next page.
Inputs: project_id (required), limit, cursor. Returns: result
list_project_labelsA
List reusable project labels. Pass next_cursor as cursor for the next page.
Inputs: limit, cursor. Returns: result
list_projectsB
List projects. Pass next_cursor as cursor for the next page.
Inputs: limit, cursor. Returns: result
list_project_updatesB
List a project's updates. Pass next_cursor as cursor for the next page.
Inputs: project_id (required), limit, cursor. Returns: result
list_teamsB
List teams. Pass next_cursor as cursor for the next page.
Inputs: limit, cursor. Returns: result
list_usersA
List organization users. Pass next_cursor as cursor for the next page.
Inputs: limit, cursor. Returns: result
list_workflow_statesC
List workflow states for a team. Pass next_cursor as cursor for the next page.
Inputs: team_id (required), limit, cursor. Returns: result
remove_issue_dependencyC
Remove an issue dependency by its relation ID.
Inputs: relation_id (required). Returns: result
remove_issue_from_projectC
Remove an issue from its current Linear project.
Inputs: issue_id (required). Returns: result
remove_issue_labelC
Remove a label from an issue.
Inputs: issue_id (required), label_id (required). Returns: result
remove_project_labelA
Remove a project label from a project without deleting the reusable label.
Inputs: label_id (required), project_id (required). Returns: result
search_issuesC
Search Linear issues with filters. Pass next_cursor as cursor for the next page.
Inputs: limit, query, state, cursor, +3 more. Returns: result
set_issue_estimateB
Set story points or an hour estimate on an issue.
Inputs: estimate (required), issue_id (required). Returns: result
set_issue_projectB
Add an issue to a Linear project.
Args: issue_id: Issue UUID or identifier such as ABS-1. project_id: Project UUID.
Inputs: issue_id (required), project_id (required). Returns: result
update_cycleA
Update a cycle's name, description, or ISO 8601 start and end timestamps.
Inputs: cycle_id (required), name, ends_at, starts_at, +1 more. Returns: result
update_issueB
Update an existing Linear issue. Use a UUID or identifier such as ART-5.
Inputs: issue_id (required), state, title, priority, +2 more. Returns: result
update_projectC
Update a project's name, summary, or status.
State accepts a Linear status type such as planned, started, paused, completed, or canceled, or the status's display name.
Inputs: project_id (required), name, state, description. Returns: result
update_project_detailsC
Update a project's detailed fields.
member_ids and team_ids are comma-separated user/team UUIDs; team keys also work. Dates use YYYY-MM-DD. Member and team lists replace their current values. Use the clear_* flags to remove an optional field deliberately.
Inputs: project_id (required), icon, color, content, +10 more. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 42 tools
Most tools map cleanly to a distinct resource-action pair, such as issue, project, cycle, label, and team. The main ambiguity is between update_project and update_project_details, whose boundaries are not immediately obvious from names alone, though their descriptions help.
Tool names overwhelmingly follow a consistent verb_noun snake_case pattern like create_issue, list_cycles, and archive_project. Minor deviations such as set_issue_project and update_project_details are understandable but slightly break the otherwise predictable convention.
With 42 tools, this server has a very large surface for agent selection. While individual tools are meaningful, the count crosses the threshold where it becomes heavy and likely to create decision overhead.
The server provides broad lifecycle coverage for issues, projects, cycles, labels, dependencies, and comments. Minor gaps exist, such as no delete operation for reusable project labels and no update operation for project updates, but core workflows are well supported.
Maintenance
Related MCP Connectors
Issues, projects, cycles, and roadmaps via the official Linear hosted MCP server.
Unofficial Linear MCP — wraps the Linear GraphQL API (OAuth)
Search, read and create Linear issues, projects, teams and cycles.
Monday.com MCP — wraps the Monday.com GraphQL API (BYO API key)
Related MCP Servers
- AlicenseNot gradedqualityFmaintenanceAn MCP server for managing Linear issues, projects, teams, and milestones through the Model Context Protocol. It enables users to create, update, and list workspace resources including issue statuses, comments, and user details.ISC
- AlicenseNot gradedqualityDmaintenanceAn MCP server for Linear that provides tools to manage issues, projects, comments, cycles, teams, and users via GraphQL, with OAuth2 authentication through Dedalus.1MIT
- AlicenseBqualityFmaintenanceMCP server for accessing and managing Linear resources like issues, projects, and comments with full CRUD operations and search.21140 npm1MIT
- AlicenseNot gradedqualityCmaintenanceCustom MCP server for Linear that enables managing issues, projects, and workspaces through natural language, backed by raw GraphQL.1MIT