Agent Skills by wlee075
Showing 10 vetted skills indexed across 1 GitHub repositories.
wlee075prompt-tuning
Modify prompts while preserving downstream regex parsing contracts.
wlee075drafting
Generate contextualized draft from prior chat Q&A history and flagged user notes.
wlee075clarification-control
Evaluate a PM response against a requirement gap and output resolved, clarification, or deferred.
wlee075elicitation
Generate one focused question per round to elicit PRD decisions.
wlee075context-assembly
Documents conditional context block assembly rules for LLM prompt nodes.
wlee075decision-normalization
Normalize PM decision responses into RESOLVED or UNRESOLVED requirement statements.
wlee075routing
Apply deterministic routing decisions to PRD graph progression after reflect and advance steps.
wlee075logging
Emit structured JSONL logs with thread_id and run_id context.
wlee075prd-writing
Consolidates system constants, state schemas, and PRD section definitions into a single reference.
wlee075reflection
Evaluate PRD section drafts against four rubrics and output verdict, score, and triage decisions.