Trello Board JSON Fetch

Why Not curl

Trello's /b/{ID}.json endpoint requires authentication. curl without cookies/tokens returns 401 for private boards. This broke our old SOP that used curl -sL "https://trello.com/b/{ID}.json".

Fetch Method: Chrome Agent

The user is logged into Trello in Chrome. Use the chrome-agent (via Task tool) to:

1. •Navigate to https://trello.com/b/{BOARD_ID}.json
2. •The page renders raw JSON (the user's session cookies authenticate)
3. •Use get_page_text to extract the full JSON content
4. •Save to the target file path

Template prompt for chrome-agent:

Navigate to https://trello.com/b/{BOARD_ID}.json in Chrome.
The user is logged into Trello so the page will show raw JSON content.

1. Call tabs_context_mcp first
2. Create a new tab navigating to https://trello.com/b/{BOARD_ID}.json
3. Wait for it to load
4. Use get_page_text to get ALL the text content (this is raw JSON)
5. Save the complete content to {TARGET_FILE_PATH}
6. Report file size and confirm save

Encoding note

Chrome-exported JSON may have UTF-8 BOM. When parsing in Python, use encoding='utf-8-sig' instead of utf-8.

File Naming Convention

{board-slug}_YYYY-MM-DD.json in the project's TrainingData/ directory.

Examples:

• •<company_a>-customer-service_2026-02-01.json
• •<company_a>-sales-training-procedures_2026-02-01.json

Fallback: Manual Browser Export

If chrome-agent fails (extension issues, tab detachment):

1. •Ask user to open https://trello.com/b/{BOARD_ID}.json in Chrome
2. •Ctrl+S to save the page as a .json file
3. •Save to the target path with dated filename

Alternative: Board UI export (Menu → ... → Print and Export → Export as JSON). Same result but through Trello UI instead of the .json URL.

Known Board Registry

This skill doesn't own any board registry. Each project maintains its own board list in its training data docs (e.g., TrainingData/RELATED_REPOSITORIES.md). Check project docs for board IDs before fetching.