kayba-ai/agentic-context-engine

Health v2 · maintenance only

Adoption not part of the score

2558 stars · 302 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Agentic Context Engine (ACE) is a Python library that lets LLM-based agents learn from experience by maintaining and evolving an evolving context/memory of strategies and lessons. It is also offered as a hosted solution that automatically investigates agent errors, proposes fixes as PRs, and tracks their performance.

Use cases

• make my ai agent learn from past mistakes
• add persistent memory to an llm agent
• improve agent prompts automatically from experience
• reduce repeated errors in my agent runs
• context engineering for llm agents
• auto-fix failing agent tool calls

When to choose

• you build LLM agents in Python and want them to accumulate learned strategies over time
• you want an open-source, self-hostable agent memory/context engine
• you want automated investigation and fix proposals for recurring agent errors

When to avoid

• you need a general-purpose agent orchestration framework rather than memory/context learning
• your stack is not Python and you cannot use an API-based integration
• you require fully deterministic, non-learning agent behavior

Facets

library · maturity active

agent-framework machine-learning llm-inference prompt-engineering large-language-models machine-learning developer-tools python cross-platform agent-memory context-engineering agent-learning llm-agents self-improving-agents ai-agents

Member repositories

For agents

markdown · JSON · MCP: product_card(name="kayba-ai/agentic-context-engine")