wewrite-stats

wewrite-stats — article data review

Execution conventions

• CLI: deterministic operations go through the wewrite command (must be in PATH; if missing, guide to uv tool install wewrite, or in the repository bash install.sh）。
• {home}: user state directory = $WEWRITE_HOME or ~/.wewrite（wewrite home can be checked). config/style/history/playbook/output/exemplars are all in {home}, not in the repository; state paths in references documents follow the same convention.
• 读取: <路径> = actually finish reading the file with a file-reading tool before continuing, not a comment.
• references/: this skill includes its own {skill_dir}/references/; the {skill_dir} inside references documents refers to this skill's directory.

Execution

读取: {skill_dir}/references/effect-review.md

Execute according to its process:fetch_stats.py --days 7 pull data → match and backfill the stats field of history.yaml → analyze best/worst performance and reasons → give adjustment suggestions for subsequent topic selection/titles/frameworks.

Prerequisite: requires the WeChat API credentials in config.yaml. Missing credentials → tell the user "data review requires configuring the Official Account API (config.yaml); currently only qualitative analysis based on existing records in history.yaml is possible", then analyze as much as possible from the existing content of history.yaml. Just-published articles → tell them to check again after 24h.

Downstream impact: the backfilled stats will be read by the topic selection module (wewrite-topic)—which framework/enhancement strategy performs well will be weighted into the next recommendation. This is half of the data closed loop; the other half is wewrite-learn's revision flywheel.