WeRead Companion

Problem

WeRead data is scattered across bookshelf, notes, reviews, and reading-statistics endpoints. Engineers often have to hand-roll pagination, JSON parsing, version checks, and redaction, which risks leaking WEREAD_API_KEY, treating personal notes as instructions, or overclaiming author identity. This skill treats the official weread-skills package as the API authority and adds only higher-level workflows and stable scripts.

How It Works

• Reading recommendations: call weread_recommend.py, then explain results by fit, mismatch, mode (safe, expand, challenge), shelf status, and next action.
• Read-before-you-commit analysis: combine book info, public reviews, popular highlights, and similar books to decide whether a book is worth starting.
• Notes and reviews: use weread_notes_export.py for Markdown/JSON exports and weread_reviews.py for public reviews and highlight thoughts, showing only API-returned author fields.
• Reading reports: use weread_report.py for weekly, monthly, annual, and bookshelf planning reports.

Boundaries

It depends on the official weread-skills skill and WEREAD_API_KEY. It is not for bypassing WeRead, identifying users beyond public fields, or presenting recommendation scores as objective facts; summaries are preferred over long quoted exports unless the user asks.