Sentry
Learning to Recover from LLM Agent Failures at Test Time
Sentry launched on 2 October 2026. It ranked #92 on Daily Launch on 22 October 2026 with a Launch Score of 24, from 1.2k stars in 20 days on GitHub. It was new in the ranking that day.
What is Sentry?
An external runtime failure management layer for LLM agents. Detect execution failures, guide recovery, and learn reusable lessons from verified recoveries. Sentry runs alongside your agent loop without replacing the agent, environment, or evaluator. Sentry is an external runtime failure management layer for LLM agents. It helps agents fix invalid actions and recover when they get stuck in loops, drift from the task, or make unsupported assumptions. Unlike typical runtime monitors with fixed advice, Sentry learns and evolves from recoveries that work.
From its README on GitHub
Sentry is a repo by nuglifeleoji on GitHub. It is written in Python and released under the MIT license.
How to get Sentry
Get it from its repository on GitHub.
Launch timeline
Launch Score, last 7 days
Ranked on one day so far: 24 of 100 on 22 October.