feat(search): read Drive documents and Gmail conversations

[waleedlatif1]

Summary

• Read text from Drive PDF and DOCX files through the existing bounded transport and parsers, with download-permission, format, page, archive and output limits.
• Read Gmail conversation context with per-message authorization, fresh permission checks, chronological attribution and explicit incomplete coverage. Normalize MIME alternatives and keep attachments separate from message bodies.
• Update query guidance and add real HTTP acceptance coverage; isolate availability tests from developer OAuth configuration.

Type of Change

• Feature

Testing

Full repository tests passed during implementation. The review corrections passed 524 provider/parser tests, 42 real HTTP acceptance checks with real PDF/DOCX parsers, 26 workspace type checks, 51 audits, lint, generators, registry and docs checks. Eight additional guard-removal controls fail independently, alongside the initial nineteen controls. Google acceptance uses synthetic HTTP fixtures; fresh connected-account verification remains untested.

Checklist

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

[vercel]

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

Project	Deployment	Actions	Updated
docs	Ready	Preview	Sep 27, 2026 2:35am UTC

[waleedlatif1]

@greptile

[waleedlatif1]

@cubic-dev-ai review this PR

[cubic-dev-ai]

@cubic-dev-ai review this PR

@waleedlatif1 I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

All reported issues were addressed across 29 files

<sub>Reply with feedback, questions, or to request a fix.

Fix all with cubic | Re-trigger cubic</sub>

[greptile-apps]

Confidence Score: 5/5

[Medium risk] Adds document text extraction from Google Drive and Gmail.

The PR appears safe to merge, though PDF reads that exceed Drive’s text limit may do avoidable parsing work.

Findings

1.  PDF budget checked too late ▶

Summary

The PR adds bounded Drive PDF/DOCX text reads and independently authorized Gmail conversation context, with real-HTTP synthetic acceptance coverage. Subsequent changes tighten DOCX conversion and correct PDF output-byte accounting. PDF reads still defer the caller’s smaller byte-budget check until after potentially substantial extraction work.

Diagram

%%{init: {'theme': 'neutral'}}%%
flowchart LR
    A[Drive file metadata] --> B[Bounded media download]
    B --> C{PDF or DOCX}
    C -->|PDF| D[Extract pages]
    D --> E[Check rendered output bytes]
    C -->|DOCX| F[Bound conversion and extract text]
    E --> G[Live document]
    F --> G
    H[Gmail anchor] --> I[Thread metadata]
    I --> J[Verify each sibling before body fetch]
    J --> K[Fresh checks before conversation output]

Loading

_{Reviews (2) · Last reviewed commit: "fix(search): bound document conversion a..."}

[waleedlatif1]

@cubic-dev-ai review this PR

[cubic-dev-ai]

@cubic-dev-ai review this PR

@waleedlatif1 I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

All reported issues were addressed across 29 files

<sub>Reply with feedback, questions, or to request a fix.

Fix all with cubic | Re-trigger cubic</sub>

[waleedlatif1]

@greptile

[waleedlatif1]

@cubic-dev-ai review this PR

[cubic-dev-ai]

@cubic-dev-ai review this PR

@waleedlatif1 I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

No issues found across 30 files

Confidence score: 5/5

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

_{Re-trigger cubic}