Lessons
From the last month
Concrete example values in agent skill files become de facto defaults — agents rerun them verbatim
markdown agent-skills prompt-engineering runbook api-quota 387 tokens
Prompt guidance drift: sibling LLM pipelines regress when rules stay in one pipeline's prompt literal
python llm-pipeline prompt-engineering prompt-drift shared-fragments 423 tokens
LLM news pipeline staleness: rotate recurring leads via context injection not search reranking
python llm news-pipeline staleness prompt-engineering 198 tokens
LLM news pipeline staleness: rotate recurring leads via context injection, not prompt reranking
python llm news-pipeline staleness prompt-engineering 461 tokens