Wink: Recovering from Misbehaviors in Coding Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908844344475648 |
|---|---|
| author | Nanda, Rahul Maddila, Chandra Jha, Smriti Khan, Euna Mehnaz Paltenghi, Matteo Chandra, Satish |
| author_facet | Nanda, Rahul Maddila, Chandra Jha, Smriti Khan, Euna Mehnaz Paltenghi, Matteo Chandra, Satish |
| contents | Autonomous coding agents, powered by large language models (LLMs), are increasingly being adopted in the software industry to automate complex engineering tasks. However, these agents are prone to a wide range of misbehaviors, such as deviating from the user's instructions, getting stuck in repetitive loops, or failing to use tools correctly. These failures disrupt the development workflow and often require resource-intensive manual intervention. In this paper, we present a system for automatically recovering from agentic misbehaviors at scale. We first introduce a taxonomy of misbehaviors grounded in an analysis of production traffic, identifying three primary categories: Specification Drift, Reasoning Problems, and Tool Call Failures, which we find occur in about 30% of all agent trajectories.
To address these issues, we developed a lightweight, asynchronous self-intervention system named Wink. Wink observes agent trajectories and provides targeted course-correction guidance to nudge the agent back to a productive path. We evaluated our system on over 10,000 real world agent trajectories and found that it successfully resolves 90% of the misbehaviors that require a single intervention. Furthermore, a live A/B test in our production environment demonstrated that our system leads to a statistically significant reduction in Tool Call Failures, Tokens per Session and Engineer Interventions per Session. We present our experience designing and deploying this system, offering insights into the challenges of building resilient agentic systems at scale. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_17037 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Wink: Recovering from Misbehaviors in Coding Agents Nanda, Rahul Maddila, Chandra Jha, Smriti Khan, Euna Mehnaz Paltenghi, Matteo Chandra, Satish Software Engineering Artificial Intelligence Human-Computer Interaction Programming Languages Autonomous coding agents, powered by large language models (LLMs), are increasingly being adopted in the software industry to automate complex engineering tasks. However, these agents are prone to a wide range of misbehaviors, such as deviating from the user's instructions, getting stuck in repetitive loops, or failing to use tools correctly. These failures disrupt the development workflow and often require resource-intensive manual intervention. In this paper, we present a system for automatically recovering from agentic misbehaviors at scale. We first introduce a taxonomy of misbehaviors grounded in an analysis of production traffic, identifying three primary categories: Specification Drift, Reasoning Problems, and Tool Call Failures, which we find occur in about 30% of all agent trajectories. To address these issues, we developed a lightweight, asynchronous self-intervention system named Wink. Wink observes agent trajectories and provides targeted course-correction guidance to nudge the agent back to a productive path. We evaluated our system on over 10,000 real world agent trajectories and found that it successfully resolves 90% of the misbehaviors that require a single intervention. Furthermore, a live A/B test in our production environment demonstrated that our system leads to a statistically significant reduction in Tool Call Failures, Tokens per Session and Engineer Interventions per Session. We present our experience designing and deploying this system, offering insights into the challenges of building resilient agentic systems at scale. |
| title | Wink: Recovering from Misbehaviors in Coding Agents |
| topic | Software Engineering Artificial Intelligence Human-Computer Interaction Programming Languages |
| url | https://arxiv.org/abs/2602.17037 |