introspection

Self-assessment applies taxonomy-driven introspection rules to automate session-end analysis, ensuring consistent and traceable memory recording. This framework guides post-mortem reviews across research workflows and LLMem integrations with auditable steps.

Structured introspection is an operational framework that applies taxonomy alignment and traceable memory recording to detail actionable rules for self-review. It standardizes error-pattern analysis by defining explicit categories, checks, and safe recording commands.

Yes, taxonomy-guided recording externalizes memory by requiring explicit taxonomy alignment during reflective work. This ensures safe, auditable steps and defined recording commands are used to capture session-end checks within LLMem integrations.

No specific dependencies or components are required to run the introspection framework. It operates as a standalone operational guide loaded during session-end to enable automatic introspection questions and taxonomy-guided recordings.

Post-mortem reviews should be triggered at session-end or during reflective work to analyze error patterns. The framework applies sampajanna checks and taxonomy-driven recording to ensure safe, auditable post-mortem analysis.

The best way to ensure traceable memory recording is to apply explicit taxonomy alignment and defined recording commands during self-assessment. This operational framework standardizes memory externalization with auditable steps and defined categories.