Self-improvement layer with evaluation separation, rollback, and tiered operator gates. Observes outcomes across sessions, detects recurring patterns, propos...
openclaw-reflect 1.0.2 - Added support for multiple evaluator backends (Anthropic, OpenAI, Ollama, rules) via new optional environment variables in SKILL.md. - Updated documentation to describe configuration using ANTHROPIC_API_KEY, OPENAI_API_KEY, OLLAMA_HOST/PORT, REFLECT_EVAL_MODEL, and REFLECT_EVALUATOR. - No changes to workflow or operator commands; improvements center on flexible evaluation options.