How is this different from connecting Shortcut MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Shortcut MCP API changes?

We track the Shortcut MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Shortcut tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Shortcut?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Shortcut MCP server myself?

No. TrustGate connects to the Shortcut MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Shortcut directly.

Connect Shortcut MCP to your AI agents, securely.

Q: Is Shortcut MCP secure?

The server itself exposes 56 tools, and some of them can write to your Shortcut account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Shortcut.

One secured endpoint for every Shortcut call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Shortcut MCP now

What your agents can ask once Shortcut MCP is connected.

Select the tools your AI agents can call

56 tools from the Shortcut MCP schema.

Showing 56 tools

stories-get-by-idreadGets a single Shortcut story by ID.
stories-searchreadFinds Shortcut stories with filtering and search options.
stories-get-historyreadGets the change history for a story.
stories-get-branch-namereadGets the recommended branch name for a story, based on workspace settings.
stories-get-by-external-linkreadFinds all stories that contain a specific external link.
stories-createwriteCreates a new Shortcut story.
stories-updatewriteUpdates an existing Shortcut story.
stories-upload-filewriteUploads a file and links it to a story.
stories-assign-current-userwriteAssigns the current user as the owner of a story.
stories-unassign-current-userwriteUnassigns the current user as the owner of a story.
stories-create-commentwriteCreates a comment on a story, which the team can see.
stories-create-subtaskwriteAdds a new sub-task to a story.
stories-add-subtaskwriteAdds an existing story as a sub-task of another story.
stories-add-taskwriteAdds a task to a story.
stories-update-taskwriteUpdates a task in a story.
stories-add-relationwriteAdds a story relationship such as relates to, blocks or duplicates, which changes what the team reads as blocked.
stories-add-external-linkwriteAdds an external link to a Shortcut story.
stories-remove-subtaskhigh riskRemoves a sub-task from a story.
stories-remove-external-linkhigh riskRemoves an external link from a Shortcut story.
stories-set-external-linkshigh riskReplaces all external links on a story with a new set, so any link omitted from the list is dropped.
iterations-get-by-idreadGets a Shortcut iteration by ID.
iterations-get-storiesreadGets the stories in a specific iteration by iteration ID.
iterations-searchreadFinds Shortcut iterations with filtering and search options.
iterations-get-activereadGets the active iterations for the current user, based on their team memberships.
iterations-get-upcomingreadGets the upcoming iterations for the current user, based on their team memberships.
iterations-createwriteCreates a new Shortcut iteration with start and end dates.
iterations-updatewriteUpdates an existing Shortcut iteration.
iterations-deletehigh riskDeletes a Shortcut iteration, removing the container the team's current work is organised under.
epics-get-by-idreadGets a Shortcut epic by ID.
epics-searchreadFinds Shortcut epics with filtering and search options.
epics-createwriteCreates a new Shortcut epic.
epics-updatewriteUpdates an existing Shortcut epic.
epics-create-commentwriteCreates a comment on an epic, which the team can see.
epics-deletehigh riskDeletes a Shortcut epic, removing the grouping that stories hang off.
documents-listreadLists all documents in Shortcut.
documents-searchreadSearches for documents.
documents-get-by-idreadRetrieves a specific document in Markdown format by its ID.
documents-createwriteCreates a new document in Shortcut with Markdown content.
documents-updatehigh riskUpdates the content of an existing document by its ID, replacing what colleagues are working from with no draft or suggestion step.
labels-listreadLists all labels in the Shortcut workspace.
labels-get-storiesreadGets all stories carrying a specific label.
labels-createwriteCreates a new label in Shortcut, adding to a shared vocabulary the whole workspace sees.
projects-listreadLists all projects in the Shortcut workspace.
projects-get-by-idreadGets a Shortcut project by its public ID.
projects-get-storiesreadGets all stories in a specific project.
workflows-listreadLists all Shortcut workflows.
workflows-get-by-idreadGets a Shortcut workflow by ID.
workflows-get-defaultreadGets the default workflow for a specific team, or the workspace default.
users-get-currentreadGets the current user's information.
users-get-current-teamsreadGets the list of teams the current user is a member of.
users-listreadGets all workspace users.
objectives-get-by-idreadGets a Shortcut objective by ID.
objectives-searchreadFinds Shortcut objectives with filtering and search options.
teams-get-by-idreadGets a Shortcut team by ID.
teams-listreadLists all Shortcut teams.
custom-fields-listreadLists all custom fields in the workspace along with their possible values.

What Shortcut MCP exposes

Shortcut MCP groups its 56 tools into 11 categories, each with a suggested starting policy.

All 56 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Stories	Reading a story, its change history and its recommended branch name, searching with filters, finding stories by external link, creating and updating stories, uploading and linking a file, assigning or unassigning the current user, commenting, managing subtasks and tasks, adding relationships such as blocks or duplicates, and adding, removing or wholesale replacing external links	Read / Write	stories-set-external-links replaces every link on a story rather than appending, so an incomplete list silently drops the rest. stories-upload-file puts a file into the workspace, and stories-add-relation changes what the team reads as blocked
Iterations	Reading an iteration and the stories in it, searching iterations, listing the active and upcoming ones for the current user based on team membership, creating an iteration with start and end dates, updating one, and deleting one	Read / Write	iterations-delete removes the container the team's current work is organised under. The active and upcoming reads are scoped to the caller's team memberships, which makes them a sensible default allow
Epics	Reading an epic by ID, searching epics with filters, creating and updating an epic, commenting on one, and deleting one	Read / Write	epics-delete removes the grouping that stories hang off. Comments are visible to the team, so treat epics-create-comment as communication rather than as a record change
Documents	Listing and searching documents, retrieving one in Markdown by ID, creating a new document with Markdown content, and updating the content of an existing one	Read / Write	documents-update replaces the content of a document colleagues work from, with no draft or suggestion step in between. Read the document first, since an update against a stale copy overwrites whatever changed since
Labels	Listing all labels in the workspace, getting every story carrying a specific label, and creating a new label	Read / Write	labels-get-stories is a useful cross-cutting read. Creating a label adds to a shared vocabulary the whole workspace sees, so it is worth keeping an eye on label sprawl rather than on risk
Projects	Listing all projects in the workspace, reading one by its public ID, and getting all stories in a specific project	Read	Read-only and safe to allow freely. Nothing here creates, changes or removes a project
Workflows	Listing all workflows, reading one by ID, and getting the default workflow for a specific team or for the workspace	Read	Read-only. Workflow states are what story updates must target, so these are the prerequisite for a valid stories-update rather than a risk of their own
Users	Getting the current user, the teams the current user belongs to, and the full list of workspace users	Read	Read-only. users-list returns the workspace roster, which is personal data about colleagues even though it is routine inside the product
Objectives	Reading an objective by ID and searching objectives with filtering options	Read	Read-only, with no way to create or change an objective through this server, so the organization's stated goals cannot be edited from an agent
Teams	Reading a team by ID and listing all teams in the workspace	Read	Read-only and a reasonable default allow. Team membership is what the iteration and workflow tools scope against
Custom Fields	Listing all custom fields in the workspace along with their possible values	Read	Read-only, and the prerequisite for setting a custom field correctly on a story. The possible values it returns are what stories-update must choose among

Connect Shortcut MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect Shortcut
Find Shortcut in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow Shortcut MCP or block it. You allow the exact tools each agent can call, one by one.

Shortcut MCP	56 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 56 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is Shortcut MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	Shortcut MCP56 tools, available as a hosted server at mcp.shortcut.com over OAuth with no API token needed, or run locally from the @shortcut/mcp package with a token in the environment. Two server-side controls narrow the surface: SHORTCUT_TOOLS takes a comma-separated allowlist that accepts either an entity name such as stories or a full tool name such as iterations-create, and SHORTCUT_READONLY set to true disables every tool that modifies data.
Coding assistants	Connection & credentialsper user	Stories20 tools
Internal copilots	Tool access policiesdeny by default	Iterations8 tools
Workflow automations	Request & response inspectioninline	Epics6 tools
Compatible MCP clients	Activity monitoringalerts and logs	All other tools22 tools