agentic-context-engine vs harness2
Learning loop for any agent: reflect on failures, distill strategies into a Skillbook, inject them next run — 2x consistency on Tau2, 49% token cuts. LiteLLM-based, 100+ providers. — versus — Google Research's recursive harnessing framework: an agent refines its own harness (prompts, tools, memory, skills) per task via Propose–Probe–Compose, with model weights fixed.
Both improve an agent without changing weights by learning from its own runs. ACE distills a skillbook; Harness² edits the whole harness.
| agentic-context-engine | harness2 | |
|---|---|---|
| Stars | 2.6k | 23k |
| Forks | 311 | 2.3k |
| Language | Python | Python |
| License | Apache-2.0 | Apache-2.0 |
| Last activity | 1 months ago | 4 days ago |
| Topics | memory, agents | agents, evals |
| Curated connections | 9 | 3 |
agentic-context-engine — the curator's take
The in-process answer to 'my agent repeats the same mistakes': wrap your agent, feed it corrections, and ACE extracts reusable strategies it injects on later runs — no fine-tuning, no reward signals, and the numbers are concrete (2x pass^4 on Tau2, ~$1.50 to learn its way through a 14k-line translation). Pick it over a memory *service* when you want the learning inside your Python process rather than behind an HTTP API. NOT magic memory: strategies come from explicit feedback loops you wire up, quality follows the judge model, and the open-source engine is the on-ramp to the hosted Kayba product — check where the managed line lands before betting infra on it.
harness2 — the curator's take
Use it if you research or tune agent harnesses and want the agent to improve its scaffolding on unseen tasks without ground-truth labels. Lessons carry forward between tasks. It is a research framework: expect paper-grade code and extra model calls per task for probing. Skip it if you want a ready coding agent or a production orchestration runtime.