zhucl1006/googleskills · Archived

google-docs-manager

Use when an AI agent needs to interact with Google Docs. Triggers when user says "create a doc", "find document", "read doc content", "append to doc", "replace text in doc", or "delete document". Requires Google Cloud Service Account credentials.

First seen Feb 27, 2026

Installation

$ npx skills add zhucl1006/googleskills --skill google-docs-manager

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from zhucl1006/googleskills.

npx skills add zhucl1006/googleskills

Browse all from zhucl1006/googleskills

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

Default branch main
Open issues 0
Status Archived

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 3,209 B
  • docs SUMMARY.md 273 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 2 installs

SKILL.md

Google Docs Manager

Manage Google Docs via Service Account — no browser interaction required, fully automatable.

Prerequisites

  1. Google Cloud project with Docs API + Drive API enabled
  2. Service Account JSON key file
  3. Set GOOGLECREDENTIALSPATH=/path/to/credentials.json
  4. Install: pip install -r requirements.txt

When to Use

  • User says: "create a doc", "write a document", "find doc", "read doc content", "append/insert/replace text in doc", "delete document"
  • Agent needs to read or modify Google Docs content programmatically
  • Not for downloading binary files or managing non-Doc files (use google-drive-manager instead)

Quick Reference

Task Command
Create doc python scripts/docs.py create "Title" [--content "text"]
Find docs python scripts/docs.py find "query" [--limit 10]
Read content python scripts/docs.py get-text <doc_id>
Append text python scripts/docs.py append-text <doc_id> "text"
Insert at start python scripts/docs.py insert-text <doc_id> "text"
Find & replace python scripts/docs.py replace-text <doc_id> "old" "new"
Delete (trash) python scripts/docs.py delete <doc_id>
Delete (permanent) python scripts/docs.py delete <doc_id> --permanent

Output Format

All commands return structured JSON:

{"status": "success", "message": "...", "data": {...}}
{"status": "error",   "message": "..."}

Agent Rules

  1. Always check status field before using data
  2. Use find before editing — never guess doc IDs
  3. Use get-text to read content before making targeted edits
  4. Prefer replace-text for targeted changes; use append-text for additions
  5. Default delete moves to trash — use --permanent only when explicitly requested
  6. docid is the alphanumeric string in the Google Docs URL: https://docs.google.com/document/d/<docid>/edit

Chained Operation Pattern

# Find → read → edit workflow
python scripts/docs.py find "meeting notes"
# → extract doc_id from data.documents[0].id

python scripts/docs.py get-text <doc_id>
# → read current content

python scripts/docs.py replace-text <doc_id> "old section" "updated section"
# → targeted edit

Auth Management

python scripts/auth.py status      # Check credentials file
python scripts/auth.py validate    # Test API connectivity

Common Mistakes

Mistake Fix
Guessing doc_id Always call find first to get exact doc_id
Editing without reading Call get-text before replace-text to confirm content exists
Permanent delete by default Keep --permanent off unless user explicitly says so
Using doc URL as doc_id Extract only the alphanumeric ID from the URL, not the full URL
Appending when replace is needed Use replace-text for targeted edits, append-text only for additions