Feature request: docs.appendMarkdown — write formatted Docs content without manual index math

[matheusrmorgado]

Problem

Producing a formatted Google Doc today takes the two-step flow the google-docs skill teaches: docs.writeText with plain text, then docs.formatText with hand-computed 1-based character indices. The skill spends most of its length teaching the model index arithmetic ("count the \n characters..."), and any miscount styles the wrong range. For agents, markdown is the native output format — the index math is pure failure surface.

Proposal

docs.appendMarkdown(documentId, markdown, tabId?): converts a markdown string to natively formatted Docs content in a single call.

• Supported: headings (#–######), **bold**, *italic*, ~~strikethrough~~, `inline code`, [links](url), bullet (- ) and numbered (1. ) lists
• One insertText with the plain text, then one batchUpdate with updateParagraphStyle / updateTextStyle / createParagraphBullets requests computed from parse offsets — no client-side index math ever reaches the model
• tabId optional, default tab when omitted (same convention as docs.writeText)
• Not in scope for v1 (documented): tables, horizontal rules, images — they degrade gracefully to plain text / are skipped with counts in the response

No new OAuth scopes (documents already covers it). Complements rather than replaces docs.formatText, which remains the right tool for fine-grained edits to existing text.

I have a working implementation (pure parser module + service method) with Jest coverage ready to submit as a PR.