Your AI tools.
Your entire workflow.

Yes. Read the outline or one section with workspace_get_document, then use workspace_edit_document_section with a heading, heading path, or block ID. Include the returned section hash so an overlapping change returns a conflict to review. Edits to other sections can be preserved without replacing the whole document. Each successful section edit saves a version under the connected person’s identity. Section edits also reach the live browser editor, where teammates can type together with named, colored cursors.

The current registered catalog contains 163 tools: 50 for Planner, 90 for QA, 13 for documents, and 10 for saved charts and reports. Your client sees the tools allowed by its scopes and module access. Individual calls also enforce permissions and applicable project restrictions. Use the explorer on this page to discover actions across all ten areas and understand the access they need.

The catalog includes tools for validating case automation, queueing automated run items, and working with supported browser and device sessions. Execution requires compatible runners, configured devices or browsers, credentials, and an available target environment. Creating a test case or recording a result alone does not execute a test. Inspect the run and its evidence before reporting success.

The setup guide explains the MCP configuration paths documented by these agent managers. A supported client connects to Leera over Streamable HTTP using OAuth or a scoped personal access token. Configure it where the agent actually runs: your Mac, a remote host, a cloud environment, or WSL. These manager-and-Leera combinations have not yet been validated end to end. The guides are not native integrations, partnerships, or connected-status indicators.

No. Leera exposes MCP over Streamable HTTP with OAuth 2.1 or personal access tokens. Clients differ in remote transport, authentication, and tool support. Follow the instructions for your client and deployment, use your instance’s actual MCP endpoint, and test a simple read before depending on a more complex workflow.

The MCP server provides a ranked keyword search across tickets, comments, ticket email, documents, test cases, test runs, test plans, and reports, plus an exact-text or regular-expression search that returns only the matching lines. Results follow the connection’s scopes and your project and document permissions. The client can then open the items it needs and read long pages in parts.

Yes. Use a token’s project allow-list and grant only the scopes the task needs. New personal access tokens default to Planner read access, with writes and deletes requiring an explicit grant. OAuth consent can request broader scopes; review them before connecting. Tool calls also respect module access, the connected person’s project permissions, and database row-level security. Set an expiry and revoke credentials when they are no longer needed.

With the required scopes and permissions, MCP tools can search, create, update, comment on, and transition issues; manage sprints; read, create, and update documents; and work with QA cases, plans, runs, results, and evidence. Project setup and saved charts and reports have their own tools and access requirements. Chart queries use supported datasets and metrics; there is no arbitrary SQL tool in the MCP catalog. Discover the tools available to your connection before starting a task.

MCP actions use the identity of the person who authorized the connection or created the token. A connected AI tool is not a separately assignable workspace member. Review the resulting issues, comments, document changes, and QA records under that person’s access.

OAuth consent and token scopes control the access you grant. Once authorized, a permitted tool call can change a record directly; Leera does not add an approval step before every mutation. Start with read access, grant only the writes you need, and review the resulting records and available history. Your AI client may have its own confirmation controls.

Leera AI is the assistant inside the workspace. Leera AI Agents provide objective and schedule configuration, manual triggers, and execution history; execution depends on the deployment, agent setup, permissions, and provider. MCP connects an external AI client to supported workspace tools. The external client runs the conversation or task; Leera supplies the scoped project access.