feat(confluence): index PowerPoint and Excel attachments

[mzxchandra]

Summary

PR #7935 indexes PDF and Word attachments on Confluence pages and blog posts as separate knowledge-base documents. PowerPoint and Excel attachments on the same pages were still dropped silently at listing time, even though the shared parser pipeline already handles them.

• Add pptx and xlsx to the Confluence attachment allowlist in connectors/confluence/attachments.ts. The set stays a deliberate Confluence-local subset of PIPELINE_PARSED_MIME_TYPES: macro-enabled, template, legacy binary (.xls, .ppt) and OpenDocument variants remain excluded.
• Update the hydration skip reason and TSDoc to name all four families.
• Update the two connector guides (knowledgebase/connectors.mdx, search/confluence.mdx) to list the new formats, the conversion advice for .xls/.ppt, and that already-connected spaces pick the formats up on their next sync.

No contentHash change is needed: ineligible attachments never produced a row, and Confluence re-lists every attachment each sync, so a newly eligible file arrives as a new document on the next run.

Type of Change

• New feature

Testing

Unit (connectors/confluence/attachments.test.ts, extended in place):

• listing includes .pptx/.xlsx stubs with the pipeline MIME types and excludes .pptm, .xls, .ppt, images and archived files
• hydration matrix covers pdf, doc, docx, pptx, xlsx
• genuine-bytes roundtrips through the public parser for PDF, DOCX, a JSZip-built PPTX and a SheetJS-built XLSX
• an attachment renamed to .ppt/.xls after listing hydrates as a replace-skip without downloading

200 tests pass across the three Confluence-related files; type-check, lint (sim and docs) and check:api-validation pass.

End to end on a local build against a Confluence Cloud test site, with a seeded page carrying Onboarding Guide.pdf, Roadmap Deck.pptx, Budget Tracker.xlsx and Old Deck.ppt:

Sync	Code	Seeded page yielded
Baseline	pre-change file	page + Onboarding Guide.pdf only
Re-sync of the same connector	this change	+ Roadmap Deck.pptx, + Budget Tracker.xlsx; .ppt still excluded (2 added, 0 deleted)

Stored objects are byte-identical to the uploaded files and keyed with their real extensions. After processing, sim knowledge search --kb returned the deck's chunk as the top hit for a sentence only the deck contains, and the spreadsheet's chunk for a sentence only the spreadsheet contains.

Reviewers: the wording in the two docs pages and the FILE_EXTENSIONS TSDoc are the judgement calls.

Checklist

• Code follows project style guidelines
• Self-reviewed my changes
• Tests added/updated and passing
• No new warnings introduced
• I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

🤖 Generated with Claude Code

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
docs	Ready	Preview	Sep 20, 2026 12:43am UTC

[mzxchandra]

@greptile @cubic review

[cubic-dev-ai]

@greptile @cubic review

@mzxchandra I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

No issues found across 4 files

Confidence score: 5/5

• Automated review surfaced no issues in the provided summaries.
• No files require special attention.

_{Re-trigger cubic}

[greptile-apps]

Confidence Score: 5/5

The PR appears safe to merge; the newly allowed formats follow established connector hydration, parsing, access, and synchronization paths with focused test coverage.

Summary

This PR extends Confluence attachment indexing to modern Excel and PowerPoint files while retaining the existing deferred-download, parser, size-limit, and permission paths.

• Adds .xlsx and .pptx to the Confluence-local attachment allowlist.
• Expands listing, hydration, genuine-file parser, and unsupported-rename tests.
• Updates both Confluence documentation surfaces with supported formats and migration guidance.

Diagram

%%{init: {'theme': 'neutral'}}%%
flowchart LR
  A[Confluence attachment listing] --> B{Supported extension?}
  B -->|pdf, doc, docx, pptx, xlsx| C[Emit deferred document stub]
  B -->|Other format| D[Exclude from listing]
  C --> E[Classify as new or changed]
  E --> F[Download original with 100 MB limit]
  F --> G[Shared parser pipeline]
  G --> H[PDF, Word, Excel, or PowerPoint parser]
  H --> I[Knowledge document processing and indexing]

Loading

_{Reviews (1) · Last reviewed commit: "feat(confluence): index PowerPoint and E..."}