Scoped Linear API key access for AI agents
Linear is increasingly used as a destination for AI-generated work: coding agents create issues from failed checks, support automations update customer-impacting bugs, and release workflows link commits to projects. Those actions can alter planning signals just as surely as source control changes alter code. Start with a dedicated integration identity and confirm the intended workspace, team, projects, issue states, labels, and workflow ownership before allowing an agent to write. A task that only needs to summarize issues or attach a link should not automatically inherit authority to create, reprioritize, or transition work across a team.
Why manage Linear credentials through Gazebo?
A shared Linear credential makes a stream of issue creation, comment changes, and state transitions difficult to connect to the agent or workflow that requested access. Linear's own identity, workspace membership, OAuth or API-token permissions, and workflow controls determine what can happen after a credential is delivered; Gazebo does not transform a broad account token into a team- or issue-limited one. Its value is making the normal retrieval path agent-specific and revocable. Review a Gazebo decision with Linear activity, issue history, workflow events, and the source run that produced the update, then remove the profile or rotate the upstream token if the agent environment is no longer trusted.
How it works
- 1
Create a dedicated Linear integration identity or token appropriate to the intended workspace and automation. Confirm team membership, project access, workflow states, and whether the task needs read-only reporting, issue creation, comments, or transitions.
- 2
Store the credential separately and create a named profile for each agent or workflow that needs it. Keep a read-only triage or reporting path separate from a write-capable issue-management path.
- 3
The approved runtime requests the mapped credential through MCP. Gazebo checks the profile before release instead of leaving the token in a repository, prompt, or shared automation setting.
- 4
Correlate the credential-retrieval event with Linear issue history, integration activity, source-control events, and the agent run. A successful retrieval is not proof that an issue was created or transitioned correctly.
- 5
Revoke the affected profile when a workflow is retired or behaves unexpectedly. If the underlying token may have escaped, remove or rotate it in Linear and review recent activity before granting replacement access.
Common use cases
Agent-attributed issue creation
A bug-detection, code-review, or monitoring agent can each have its own profile and expected destination team. Use Linear's issue history to validate the title, owner, state, labels, project, and links created by the workflow; use Gazebo's record to identify which agent was permitted to obtain the credential for that run.
Workflow automation audit trail
Zapier, n8n, and similar workflows should receive separate profiles for separate business processes. Keep trigger validation and deduplication in the workflow itself, then correlate Gazebo retrievals with the workflow run and Linear history to investigate a duplicate, unexpected status transition, or issue created from untrusted input.
Connect Linear to Gazebo
Give your agents scoped access to Linear in minutes. Every call logged. Revoke anytime.
Connect LinearAgents that commonly use Linear
Further reading
How to Set Up a Cursor Agent with Scoped Service Access
Stop putting raw API keys in your .env. Connect Cursor to Gazebo and give your agent exactly the access it needs — nothing more.
OAuth 2.0 for Agents: Why Client Credentials Aren't Enough
OAuth 2.0's client credentials grant was designed for services, not autonomous AI agents. Here's what breaks at agent scale and what a better pattern looks like.
RBAC for AI Agents: Does Role-Based Access Control Work?
RBAC works for agents — but only if you drop the assumption that makes it useful for humans. Here's what breaks, what to use instead, and how HashiCorp Vault's policy model fits in.
AI Agent Permissions: Service, Action, and Data Scope
A permission-design guide for AI agents: define service, action, and data boundaries before issuing any credential.