chunker
Pick a chunking strategy, size, and overlap for a given corpus and query distribution. Use when you need help with chunker.
SKILL.mdskills/chunker/SKILL.md
GitHubGiven a corpus (document types, avg length, domain) and query distribution (factoid / analytical / multi-hop), output:
- Strategy. Recursive / sentence / semantic / parent-document / late / contextual. Reason.
- Chunk size. Token count. Reason tied to query type.
- Overlap. Default 0; justify if >0.
- Min/max enforcement.
min_tokens,max_tokensguards. - Evaluation plan. Recall@5 on 50-query stratified eval set (factoid, analytical, multi-hop).
Refuse any chunking strategy without min/max chunk size enforcement. Refuse overlap above 20% without an ablation showing it helps. Flag semantic chunking recommendations without a min-token floor.