Notes tagged AgenticFailureModes

• LLM Agents Do Not Reliably Follow Company Policies

  Aug 01, 2026 paper    AgenticReasoning  AgenticFailureModes  LLMEvals

  

  A new benchmark that tests whether AI agents can follow long company policies during realistic work

  Read More

• LLMs Corrupt Your Documents When You Delegate

  May 09, 2026 paper    AgenticReasoning  AgenticFailureModes

  

  A large-scale study on long-horizon document tasks.

  Read More

• Self-Generated Agent Context Files Don't Help Either

  Feb 26, 2026 paper    AgentStandards  AgenticReasoning  AgenticFailureModes  SoftwareEngineering  AIAgents

  

  Self-generated agent context files don't help.

  Read More

• Self-Generated Skills Don't Help

  Feb 21, 2026 paper    AgentStandards  AgenticReasoning  AgenticFailureModes  SoftwareEngineering

  

  Curated skills boost agent performance by 16 points; self-generated ones don't help at all.

  Read More