What can @shortcut/mcp do?

shortcut project and story management via mcp protocol, story and epic querying with filtering and pagination, story and epic creation with automatic field validation, story and epic mutation (update, delete) with change tracking, workflow state and project metadata discovery, mcp protocol transport and schema mapping

@shortcut/mcp

MCP ServerFree

Shortcut MCP Server

Open Source

/ 100

6 capabilities

Capabilities6 decomposed

shortcut project and story management via mcp protocol

Medium confidence

Exposes Shortcut's project management API through the Model Context Protocol, enabling Claude and other MCP-compatible AI clients to read and manipulate stories, epics, projects, and workflows. Implements MCP resource and tool schemas that map Shortcut's REST API endpoints into standardized tool definitions, allowing LLMs to query project state and trigger mutations without direct API knowledge.

Solves for

I want Claude to understand my current Shortcut project structure and active storiesI need an AI agent to automatically create or update stories based on conversation contextI want to query story status, assignees, and estimates without leaving my AI chat interfaceI need to sync story updates from Claude back into Shortcut without manual copy-paste

Best for

engineering teams using Shortcut for project management who want AI-assisted planning

solo developers building LLM agents that need to interact with Shortcut workflows

teams migrating from manual Shortcut updates to AI-driven story creation and management

Requires

Node.js 16+ (MCP server runtime)

Valid Shortcut API token with read/write scopes

MCP-compatible client (Claude Desktop, or custom MCP client implementation)

Limitations

Requires valid Shortcut API token with appropriate scopes — no built-in token refresh or rotation

MCP protocol overhead adds ~100-200ms per request compared to direct REST calls

No built-in caching of project state — each query hits Shortcut API, risking rate limits under high-frequency agent usage

What makes it unique

Implements MCP as a bridge between Shortcut's REST API and Claude/LLM clients, using standardized MCP tool schemas to abstract Shortcut's domain model (stories, epics, workflows) into composable agent capabilities. This allows LLMs to reason about project state and trigger mutations through a protocol-agnostic interface rather than direct API calls.

vs alternatives

Provides native MCP integration for Shortcut (vs. building custom Claude plugins or REST wrappers), enabling seamless multi-turn agent workflows without context switching or manual API orchestration.

story and epic querying with filtering and pagination

Medium confidence

Implements read-only MCP tools that query Shortcut's story and epic endpoints with support for filtering by project, workflow state, assignee, and label. Uses Shortcut's query parameters to enable efficient server-side filtering, reducing payload size and API calls. Supports pagination to handle large result sets without overwhelming the LLM context window.

Solves for

I want to find all stories assigned to a specific team member in the current sprintI need to list all epics in a project and their completion statusI want to search for stories with a specific label or tagI need to retrieve stories in a particular workflow state (e.g., 'in progress', 'done')

Best for

AI agents that need to understand project scope before making decisions

teams building LLM-powered dashboards or status reports from Shortcut data

developers creating context-aware agents that must filter stories by multiple dimensions

Requires

Shortcut API token with read access to stories/epics

Knowledge of valid project IDs, workflow state names, and label names in the Shortcut workspace

Limitations

Query complexity limited by Shortcut API's filter syntax — cannot perform complex boolean queries (e.g., 'stories assigned to Alice OR Bob AND in sprint X')

Pagination requires explicit offset/limit parameters — no cursor-based pagination, making large result sets inefficient

No full-text search across story descriptions — only metadata filtering (state, assignee, labels)

What makes it unique

Exposes Shortcut's server-side filtering capabilities through MCP tool parameters, allowing the LLM to specify complex queries (by project, state, assignee, label) that are evaluated server-side rather than client-side, reducing data transfer and improving query performance.

vs alternatives

More efficient than fetching all stories and filtering in-memory because it leverages Shortcut's native query parameters, reducing API payload size and enabling agents to work with large projects without context window exhaustion.

story and epic creation with automatic field validation

Medium confidence

Implements MCP tools for creating stories and epics in Shortcut with schema-based validation of required fields (name, project, workflow state). Validates input parameters against Shortcut's API constraints before submission, providing early error feedback to the LLM. Supports optional fields like description, estimate, assignee, and labels, allowing flexible story creation workflows.

Solves for

I want Claude to create a new story in Shortcut based on a conversation or requirementI need to bulk-create stories from a list of requirements with consistent metadataI want to create an epic and automatically link child stories to itI need to create a story with a specific estimate and assignee based on context

Best for

teams using AI to generate stories from requirements or user feedback

developers building agents that automate story creation workflows

non-technical product managers who want Claude to translate ideas into Shortcut stories

Requires

Shortcut API token with write access to stories

Valid project ID where the story will be created

Valid workflow state name (e.g., 'Unstarted', 'In Progress')

Limitations

No transaction support — if story creation succeeds but label assignment fails, the story is created without labels (no rollback)

Validation is schema-level only — cannot validate business logic (e.g., 'estimate must be <= sprint capacity')

No bulk creation endpoint — multiple stories require sequential API calls, increasing latency

What makes it unique

Implements client-side schema validation before submitting to Shortcut's API, catching invalid field combinations (e.g., invalid workflow state) before making network requests. This reduces API errors and provides immediate feedback to the LLM, improving agent reliability.

vs alternatives

Validates input parameters before API submission (vs. raw REST calls that fail server-side), enabling agents to recover from validation errors without wasting API calls or creating partial/invalid stories.

story and epic mutation (update, delete) with change tracking

Medium confidence

Implements MCP tools for updating story/epic fields (name, description, state, estimate, assignee, labels) and deleting stories. Supports partial updates — only specified fields are modified, leaving others unchanged. Tracks which fields were modified and returns the updated object, enabling agents to confirm changes and detect conflicts.

Solves for

I want Claude to move a story to 'Done' when a task is completedI need to update a story's estimate based on new informationI want to reassign a story to a different team memberI need to delete a story that's no longer relevant

Best for

agents that manage story lifecycle (creation → progress → completion)

teams automating story state transitions based on external events

developers building AI-powered project management workflows

Requires

Shortcut API token with write access to stories

Valid story/epic ID to update or delete

For state transitions: valid workflow state name

Limitations

No optimistic locking — concurrent updates from multiple agents can overwrite each other without conflict detection

Partial updates are applied sequentially — if one field update fails, previous updates are not rolled back

No audit trail of who/what changed a story — Shortcut's audit log is server-side only

What makes it unique

Supports partial updates with field-level granularity, allowing agents to modify only specific story attributes without re-submitting the entire object. Returns the updated object to enable agents to verify changes and detect conflicts.

vs alternatives

More flexible than full-object replacement because it allows agents to update individual fields (e.g., just the state) without needing to fetch and re-submit the entire story, reducing API calls and enabling atomic field updates.

workflow state and project metadata discovery

Medium confidence

Implements MCP tools that query Shortcut's workflow definitions and project metadata, returning valid workflow states, project names, and team structures. Enables agents to discover valid values for constrained fields (e.g., workflow states) without hardcoding them, improving robustness across different Shortcut workspaces.

Solves for

I want Claude to know what workflow states are available in my Shortcut workspaceI need to list all projects so the agent can create stories in the right oneI want the agent to understand the team structure and valid assigneesI need to discover available labels and custom fields for story creation

Best for

agents that must adapt to different Shortcut workspace configurations

teams with custom workflows that differ from Shortcut defaults

developers building multi-workspace agents that need to discover valid values dynamically

Requires

Shortcut API token with read access to workspace metadata

Network access to Shortcut API

Limitations

Metadata discovery adds latency — agents must fetch workflow/project info before creating stories, increasing round-trip time

No caching of metadata — each agent session re-fetches the same data, wasting API calls

Metadata changes (new workflow state, new project) are not reflected until agent re-fetches

What makes it unique

Exposes Shortcut's workspace configuration as queryable MCP tools, allowing agents to dynamically discover valid workflow states, projects, and team members rather than relying on hardcoded values. This enables agents to adapt to different workspace configurations without modification.

vs alternatives

Enables dynamic configuration discovery (vs. hardcoding workflow states and project IDs), making agents portable across different Shortcut workspaces and resilient to configuration changes.

mcp protocol transport and schema mapping

Medium confidence

Implements the MCP server-side protocol handler that translates between MCP tool calls (from Claude or other MCP clients) and Shortcut API requests. Uses JSON schema definitions to describe tool parameters and return types, enabling clients to understand tool capabilities without documentation. Handles MCP request/response serialization, error handling, and protocol versioning.

Solves for

I want Claude to understand what Shortcut operations are available and their parametersI need reliable request/response handling between Claude and Shortcut APII want to use Shortcut integration with any MCP-compatible client, not just Claude

Best for

teams building multi-tool AI agents that need standardized protocol integration

developers creating custom MCP clients that need Shortcut support

organizations standardizing on MCP for LLM tool integration

Requires

Node.js 16+ runtime

MCP-compatible client (Claude Desktop, custom implementation)

Shortcut API token

Limitations

MCP protocol overhead adds ~100-200ms per request vs. direct REST calls

Schema validation is client-side — malformed requests from non-compliant clients may fail at Shortcut API level

No built-in request queuing or rate limiting — high-frequency agent usage can hit Shortcut API rate limits

What makes it unique

Implements MCP as a protocol layer that abstracts Shortcut's REST API, using JSON schemas to describe tool capabilities. This enables any MCP-compatible client (not just Claude) to interact with Shortcut through a standardized interface.

vs alternatives

Provides protocol-agnostic integration (vs. Claude-specific plugins) by implementing MCP, enabling the same Shortcut tools to work with multiple LLM clients and frameworks that support MCP.

Capabilities are decomposed by AI analysis. Each maps to specific user intents and improves with match feedback.

Related Artifactssharing capabilities

Artifacts that share capabilities with @shortcut/mcp, ranked by overlap. Discovered automatically through the match graph.

MCP Server24

Agile Luminary

** - Simpler Project Management - send [Agile Luminary](https://agileluminary.com) stories straight to your IDE

story status and metadata querying via mcp toolsmcp-based project management story synchronizationide-native story filtering and searchai-assisted story decomposition and task generation

4 shared capabilities

MCP Server31

@shortcut/mcp

Shortcut MCP Server

shortcut tool-based action execution via mcpshortcut workspace resource enumeration and filteringshortcut entity schema mapping to mcp typesshortcut workspace resource exposure via mcp protocol

4 shared capabilities

MCP Server25

Storyblok

** - Storyblok MCP server enables your AI assistants to directly access and manage your Storyblok spaces, stories, components, assets, workflows, and more.

storyblok space and story management via mcp protocolworkflow and publishing state managementbatch content operations and bulk updates

3 shared capabilities

MCP Server29

storybook-mcp-server

MCP server for Storybook - provides AI assistants access to components, stories, properties and screenshots

mcp-tool-exposure-for-ai-assistantscomponent-metadata-extraction-and-indexingstory-variant-enumeration-and-filtering

3 shared capabilities

MCP Server20

Project Manager

** - Hierarchical task management (ideas → epics → tasks) with CLI dashboard

mcp tool-use integration for ai-driven task management

1 shared capability

Product40

MidReal

Transform ideas into captivating stories with AI-driven, user-guided narrative...

story-project-organization-and-versioning

1 shared capability

Best For

✓engineering teams using Shortcut for project management who want AI-assisted planning
✓solo developers building LLM agents that need to interact with Shortcut workflows
✓teams migrating from manual Shortcut updates to AI-driven story creation and management
✓AI agents that need to understand project scope before making decisions
✓teams building LLM-powered dashboards or status reports from Shortcut data
✓developers creating context-aware agents that must filter stories by multiple dimensions
✓teams using AI to generate stories from requirements or user feedback
✓developers building agents that automate story creation workflows

Known Limitations

⚠Requires valid Shortcut API token with appropriate scopes — no built-in token refresh or rotation
⚠MCP protocol overhead adds ~100-200ms per request compared to direct REST calls
⚠No built-in caching of project state — each query hits Shortcut API, risking rate limits under high-frequency agent usage
⚠Limited to Shortcut's API capabilities — cannot perform operations unsupported by Shortcut REST API
⚠No transaction support — multi-step operations (e.g., create story + add labels + assign) require separate calls
⚠Query complexity limited by Shortcut API's filter syntax — cannot perform complex boolean queries (e.g., 'stories assigned to Alice OR Bob AND in sprint X')

Requirements

Node.js 16+ (MCP server runtime)Valid Shortcut API token with read/write scopesMCP-compatible client (Claude Desktop, or custom MCP client implementation)Network access to Shortcut API endpoints (api.app.shortcut.com)Shortcut API token with read access to stories/epicsKnowledge of valid project IDs, workflow state names, and label names in the Shortcut workspaceShortcut API token with write access to storiesValid project ID where the story will be created

Input / Output

Accepts: natural language queries (e.g., 'list all stories in the current sprint'), structured parameters (story IDs, project names, workflow states), JSON payloads for story creation/update operations, filter parameters (project_id, workflow_state, assignee_id, label_name), pagination parameters (limit, offset), story name (required, string), project ID (required, integer), workflow state (required, string), description (optional, string), estimate (optional, integer), assignee ID (optional, integer), label IDs (optional, array of integers), epic ID (optional, integer), story ID (required, integer), fields to update (optional, object with name, description, state, estimate, assignee_id, label_ids), delete flag (optional, boolean), optional: project ID to filter workflows for that project, MCP tool call requests with JSON parameters

Produces: structured JSON (stories, projects, epics with metadata), natural language summaries of project state, confirmation responses for mutations (create/update/delete), JSON array of story objects with metadata (id, name, state, assignee, estimate, labels), JSON array of epic objects with child story counts and completion percentages, created story object with ID, name, state, and metadata, error response with validation details if creation fails, updated story object with new values, confirmation of deletion (if delete operation), error response if update fails (e.g., invalid state transition), JSON array of workflow states with names and IDs, JSON array of projects with names and IDs, JSON array of team members with names and IDs, JSON array of available labels, MCP tool response with JSON result or error

UnfragileRank

Adoption47%(25% weight)

Quality12%(25% weight)

Ecosystem49%(15% weight)

Match Graph25%(30% weight)

Freshness75%(5% weight)

UnfragileRank is computed from adoption signals, documentation quality, ecosystem connectivity, match graph feedback, and freshness. No artifact can pay for a higher rank.

Type: MCP Server

6 capabilities

Visit @shortcut/mcp→

Repository Details

Package Details

npm

Registry

0.24.0

Version

11,318

Weekly Downloads

About

Shortcut MCP Server

Alternatives to @shortcut/mcp

LangChain72Framework

Revolutionize AI application development, monitoring, and...

Compare →

Bubble AI71Product

No-code AI app builder from natural language.

Compare →

LlamaIndex70Framework

Transform enterprise data into powerful LLM applications...

Compare →

Glide70Product

No-code app builder from spreadsheets — AI-generated mobile and web apps.

Compare →

Are you the builder of @shortcut/mcp?

Claim this artifact to get a verified badge, access match analytics, see which intents users search for, and manage your listing.

Claim this artifact →Verification via email

Get the weekly brief

New tools, rising stars, and what's actually worth your time. No spam.

Data Sources

npm

Looking for something else?

Search →

Capabilities6 decomposed

shortcut project and story management via mcp protocol

Medium confidence

Solves for

Best for

engineering teams using Shortcut for project management who want AI-assisted planning

solo developers building LLM agents that need to interact with Shortcut workflows

teams migrating from manual Shortcut updates to AI-driven story creation and management

Requires

Node.js 16+ (MCP server runtime)

Valid Shortcut API token with read/write scopes

MCP-compatible client (Claude Desktop, or custom MCP client implementation)

Limitations

Requires valid Shortcut API token with appropriate scopes — no built-in token refresh or rotation

MCP protocol overhead adds ~100-200ms per request compared to direct REST calls

No built-in caching of project state — each query hits Shortcut API, risking rate limits under high-frequency agent usage

What makes it unique

vs alternatives

Provides native MCP integration for Shortcut (vs. building custom Claude plugins or REST wrappers), enabling seamless multi-turn agent workflows without context switching or manual API orchestration.

story and epic querying with filtering and pagination

Medium confidence

Solves for

Best for

AI agents that need to understand project scope before making decisions

teams building LLM-powered dashboards or status reports from Shortcut data

developers creating context-aware agents that must filter stories by multiple dimensions

Requires

Shortcut API token with read access to stories/epics

Knowledge of valid project IDs, workflow state names, and label names in the Shortcut workspace

Limitations

Query complexity limited by Shortcut API's filter syntax — cannot perform complex boolean queries (e.g., 'stories assigned to Alice OR Bob AND in sprint X')

Pagination requires explicit offset/limit parameters — no cursor-based pagination, making large result sets inefficient

No full-text search across story descriptions — only metadata filtering (state, assignee, labels)

What makes it unique

vs alternatives

story and epic creation with automatic field validation

Medium confidence

Solves for

Best for

teams using AI to generate stories from requirements or user feedback

developers building agents that automate story creation workflows

non-technical product managers who want Claude to translate ideas into Shortcut stories

Requires

Shortcut API token with write access to stories

Valid project ID where the story will be created

Valid workflow state name (e.g., 'Unstarted', 'In Progress')

Limitations

No transaction support — if story creation succeeds but label assignment fails, the story is created without labels (no rollback)

Validation is schema-level only — cannot validate business logic (e.g., 'estimate must be <= sprint capacity')

No bulk creation endpoint — multiple stories require sequential API calls, increasing latency

What makes it unique

vs alternatives

story and epic mutation (update, delete) with change tracking

Medium confidence

Solves for

Best for

agents that manage story lifecycle (creation → progress → completion)

teams automating story state transitions based on external events

developers building AI-powered project management workflows

Requires

Shortcut API token with write access to stories

Valid story/epic ID to update or delete

For state transitions: valid workflow state name

Limitations

No optimistic locking — concurrent updates from multiple agents can overwrite each other without conflict detection

Partial updates are applied sequentially — if one field update fails, previous updates are not rolled back

No audit trail of who/what changed a story — Shortcut's audit log is server-side only

What makes it unique

vs alternatives

workflow state and project metadata discovery

Medium confidence

Solves for

Best for

agents that must adapt to different Shortcut workspace configurations

teams with custom workflows that differ from Shortcut defaults

developers building multi-workspace agents that need to discover valid values dynamically

Requires

Shortcut API token with read access to workspace metadata

Network access to Shortcut API

Limitations

Metadata discovery adds latency — agents must fetch workflow/project info before creating stories, increasing round-trip time

No caching of metadata — each agent session re-fetches the same data, wasting API calls

Metadata changes (new workflow state, new project) are not reflected until agent re-fetches

What makes it unique

vs alternatives

Enables dynamic configuration discovery (vs. hardcoding workflow states and project IDs), making agents portable across different Shortcut workspaces and resilient to configuration changes.

mcp protocol transport and schema mapping

Medium confidence

Solves for

Best for

teams building multi-tool AI agents that need standardized protocol integration

developers creating custom MCP clients that need Shortcut support

organizations standardizing on MCP for LLM tool integration

Requires

Node.js 16+ runtime

MCP-compatible client (Claude Desktop, custom implementation)

Shortcut API token

Limitations

MCP protocol overhead adds ~100-200ms per request vs. direct REST calls

Schema validation is client-side — malformed requests from non-compliant clients may fail at Shortcut API level

No built-in request queuing or rate limiting — high-frequency agent usage can hit Shortcut API rate limits

What makes it unique

vs alternatives

Provides protocol-agnostic integration (vs. Claude-specific plugins) by implementing MCP, enabling the same Shortcut tools to work with multiple LLM clients and frameworks that support MCP.

Capabilities are decomposed by AI analysis. Each maps to specific user intents and improves with match feedback.

Alternatives to @shortcut/mcp

LangChain72Framework

Revolutionize AI application development, monitoring, and...

Compare →

Bubble AI71Product

No-code AI app builder from natural language.

Compare →

LlamaIndex70Framework

Transform enterprise data into powerful LLM applications...

Compare →

Glide70Product

No-code app builder from spreadsheets — AI-generated mobile and web apps.

Compare →

@shortcut/mcp

Capabilities6 decomposed

shortcut project and story management via mcp protocol

story and epic querying with filtering and pagination

story and epic creation with automatic field validation

story and epic mutation (update, delete) with change tracking

workflow state and project metadata discovery

mcp protocol transport and schema mapping

Related Artifactssharing capabilities

Agile Luminary

@shortcut/mcp

Storyblok

storybook-mcp-server

Project Manager

MidReal

Best For

Known Limitations

Requirements

Input / Output

UnfragileRank

Repository Details

Package Details

About

Categories

Alternatives to @shortcut/mcp

Are you the builder of @shortcut/mcp?

Get the weekly brief

Data Sources

@shortcut/mcp

Capabilities6 decomposed

shortcut project and story management via mcp protocol

story and epic querying with filtering and pagination

story and epic creation with automatic field validation

story and epic mutation (update, delete) with change tracking

workflow state and project metadata discovery

mcp protocol transport and schema mapping

Related Artifactssharing capabilities

Agile Luminary

@shortcut/mcp

Storyblok

storybook-mcp-server

Project Manager

MidReal

Best For

Known Limitations

Requirements

Input / Output

UnfragileRank

Repository Details

Package Details

About

Categories

Alternatives to @shortcut/mcp

Are you the builder of @shortcut/mcp?

Get the weekly brief

Data Sources