learnings-researcher

To search bilingual solution articles, this Skill extracts relevant documented learnings from English and Chinese docs, treating counterparts as a single learning. It ranks results by relevance using frontmatter fields like module, problem_type, and symptoms to prevent rework.

YAML frontmatter drives ranking by parsing module, problem_type, component, symptoms, root_cause, tags, and severity fields. These metadata fields prioritize solution articles relevant to your current development work, returning concise, actionable summaries with citations.

Retrieving prior learnings from a bilingual knowledge base involves searching docs/en/solutions and docs/zh-Hans/solutions directories. The Skill fuses English and Chinese counterparts into a single learning, ranking them by frontmatter relevance to inform feature planning.

Yes, extracting actionable summaries treats bilingual counterparts as a single learning. The Skill searches both English and Chinese solution docs, fuses duplicate counterparts, and ranks results by YAML frontmatter fields to produce concise, cited summaries.

Ranking bilingual learnings requires solution articles within docs/en/solutions and docs/zh-Hans/solutions directories. The Skill parses YAML frontmatter fields including module, problem_type, component, symptoms, root_cause, tags, and severity from these specific paths to rank results.

Bilingual learnings retrieval is limited to documented solutions within the specified docs/en/solutions and docs/zh-Hans/solutions paths. If prior learnings are undocumented or lack structured YAML frontmatter, ranking and extraction cannot return actionable summaries.