@DanKornas: Agents shouldn’t make the same mistake every session. Agentic Context Engine (ACE) is an open-source Python engine for …

Agents shouldn’t make the same mistake every session. Agentic Context Engine (ACE) is an open-source Python engine for adding a persistent learning loop to AI agents. It helps agents reuse lessons from feedback and traces by maintaining a Skillbook: a collection of strategies that gets updated as tasks run, fail, and improve. Key features: • Persistent Skillbook – stores strategies that evolve across tasks instead of disappearing after one session • Reflection loop – a Reflector analyzes execution traces while a SkillManager adds, refines, and removes strategies • Simple quick start – install with uv, run ace setup, then use ACELiteLLM to ask, correct, and inspect learned strategies • Trace-based learning – extract strategies from existing agent logs without re-running the original tasks • Runner integrations – includes paths for LiteLLM, browser-use, LangChain, Claude Code, and MCP workflows It’s open-source under the Apache License 2.0. Link in the reply