Pinned Loading

1. AMA-Bench/AMA-Bench AMA-Bench/AMA-Bench Public

   [ICML 26] An evaluation framework assessing long-context retention and long-horizon memory performance for agentic applications (AMA-bench).

   Python 83 12

2. sysevol-ai/CodeNib sysevol-ai/CodeNib Public

   Find source context and trace calls across languages in Claude Code, Codex and other MCP agents.

   Python 86 4

3. memory-probe memory-probe Public

   [ICLR '26 W] Diagnosing Retrieval vs. Utilization Bottlenecks in LLM Agent Memory https://arxiv.org/abs/2603.02473

   Python 17 4

4. benchflow-ai/skillsbench benchflow-ai/skillsbench Public

   SkillsBench evaluates how well skills work and how effective agents are at using them.

   PDDL 1.8k 370