Montra

By Montra

Get New App Alerts

Open in ChatGPT

Open in Claude

Overview

Montra turns narrated screen recordings into structured tickets with annotated screenshots and transcripts; this connector lets Claude browse your workspace's documents and tickets, pull the screenshots and transcripts behind each one, and create, update, link, and close tickets as it implements the work.

Tools

create_document

ChatGPT

Create a draft Montra document in the connected workspace. Only call this when the user explicitly asks for a new document in this conversation — never on your own initiative.

create_ticket

ChatGPT

Add a ticket to a Montra document. Only call this when the user explicitly asks you to add a ticket in this conversation — never on your own initiative, e.g. to break work down or log follow-ups. body is the ticket's markdown implementation spec; summary is the one-line gist shown next to the title.

delete_ticket

ChatGPT

Remove a ticket from its Montra document. Only call this when the user explicitly asks you to delete that ticket — never to clean up work you consider finished, duplicated, or out of scope. The row is soft-deleted, so it stops appearing in the document but stays recoverable from the app.

get_document

ChatGPT

Read one Montra document with its folder, its tickets (each ticket body is the markdown implementation spec), and the status of its latest generation.

get_document_by_share_url

ChatGPT

Resolve a pasted Montra share link (montra.com/share/<token> or /tasks-for-agents/<token>) into its full machine-readable payload — document, tickets with markdown specs, annotated screenshot URLs, recordings, and agent instructions. Always use this instead of fetching or browsing the share web page. Works on any workspace, and on invite-only links too: when the link is not publicly readable it resolves as the Montra user who authorized this connection.

get_ticket

ChatGPT

Read one Montra ticket in full, including body — its markdown implementation spec. Document and ticket lists omit body, so fetch the ticket here before implementing it.

get_ticket_images

ChatGPT

Get the screenshots attached to a Montra ticket — real UI captures the person marked up while recording, so read them as part of the spec rather than as decoration. Each url is a short-lived signed link; fetch the images while you are working the ticket instead of storing the URLs.

get_transcript

ChatGPT

Read what the person said while recording, as the narration behind a document's tickets. Pass either a documentId for every transcribed input on the document, or a single inputId. This is supporting context: the tickets remain the spec.

get_workspace

ChatGPT

Show which Montra workspace this connection is bound to.

link_ticket_issue

ChatGPT

Record where a Montra ticket now lives downstream — e.g. the Linear issue you just created from it — so the ticket shows a link back. Whenever the user asks you to create an external issue from a ticket, record it here as part of that export — no separate ask is needed. Do not call it for issues you did not create, and it is not a way to mark a ticket done. One link per ticket and provider; linking again re-points the existing one.

list_documents

ChatGPT

List recent Montra documents in the connected workspace, optionally limited to one folder.

list_folders

ChatGPT

List the folders in the connected workspace. Use a folder id to narrow list_documents.

search_workspace

ChatGPT

Semantic search across the connected workspace's documents, tickets, transcripts, inputs, analyses, and folders. Hits are resolved to the real entity they belong to: each result is a ticket, document, or folder with its own title and verbatim body text. Results are ranked by relevance and limited to what the connected user can access.

update_document

ChatGPT

Replace the title and/or summary of a Montra document. Only call this when the user explicitly asks you to change them in this conversation — never on your own initiative, and never to note progress. Overwrites the stored values, including edits made by people.

update_ticket

ChatGPT

Replace a Montra ticket's title, summary, or markdown body. Only call this when the user explicitly asks you to change the ticket's text in this conversation — never on your own initiative, and never to record progress (e.g. adding 'implemented', 'done', or '✅' to the title or body). The ticket is the person's spec; report progress in conversation instead. Each field you pass overwrites the stored value outright, so send the complete new text rather than a fragment or a diff.