edit_comment

Replace the text of one of your own comments with Markdown. Only the author can edit a comment, and needs edit access to the project. The previous text is kept in the comment's version history. The id is shown by get_issue_details.

• 2026-10-06 — A sysdraw diagram can be embedded in a description or comment: write its view link as an image line,  (a bare link stays a link); get_issue_details shows it the same way. Only view links anyone can open are accepted; a comment, edit or signed-in-only link is refused with the reason, and a new diagram is saved only after sysdraw confirmed it (a person can add about 60 new links per 15 minutes; when sysdraw cannot be asked the call fails and can be retried). Diagrams already in the text are not checked again.
• 2026-10-09 — A Mermaid diagram can be written in a description or comment as a fenced code block: ```mermaid, the diagram source, ```. The web app draws it as a diagram (with a Show source toggle); the source stays in the text and get_issue_details gives it back unchanged. Not drawn (the source is shown instead): config directives (%%{init: ...}%% lines), front matter (a leading --- block), image shapes (img:), and diagrams over about 20 000 characters or 500 lines; one over about 8 000 characters or 150 lines (a mindmap: 80 lines) is drawn when the reader presses Draw.
• 2026-10-03 — Also answers structuredContent (see outputSchema); a failed call adds structuredContent.error. Text unchanged. Details: /docs/mcp, "Structured results and errors".
• 2026-10-04 — Claude Code 2.1.288+ (clientInfo claude-code): the text of a successful answer is itself up to 512 bytes, else "Result in structuredContent (N bytes)."; structuredContent and errors are unchanged. See /docs/mcp.