Skip to content

Playbook

Distill a trace

Compile a paid-for lesson into a skill, a rubric, or a golden case. Do not hoard transcripts.

When: The same failure happened twice, or a run was surprisingly good.

  1. 01

    Open the trace

    Langfuse (or equivalent). Name the missing rule, the bad schema, or the missing eval.

    Observability
  2. 02

    Write the artifact

    Skill, promptfoo case, or DSPy metric. Smallest artifact that would have prevented the miss.

    Distill Traces into Skills
  3. 03

    Do not store the novel

    Memory writes are curated tool calls. Session logs are not memory.

    Long-Term Memory

Done when

  • A new eval case and/or a SKILL.md, linked from the trace. Raw transcript is not 'memory'.

Refuse

  • Memory as unfiltered logs. Memory writes are tool calls. Curate. Letta's core vs archival. User can inspect/delete.
  • Eval by demo. 20–50 real tasks before the framework. promptfoo, Inspect, DSPy. CI on prompt change.