{
  "id": 7242363,
  "title": "When Someone Else's System Rejects You Without a Reason — Fixing a Diagnostic Order in Advance",
  "url": "https://urgent.news/2026/09/14/when-someone-elses-system-rejects-you-without-a-reason-fixing-a",
  "topic": "health",
  "section": "Health & Medicine",
  "published": "2026-09-14T04:10:06.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/dbsoul/when-someone-elses-system-rejects-you-without-a-reason-fixing-a-diagnostic-order-in-advance-4n5b"
  },
  "original_language": "en",
  "account": "This is chapter 11 from the book Building Autonomous AI Agents with Claude Code. It discusses a situation where an automation failed due to an unspecified error and how to prevent it. The author built an automation that uploads a submission file to an external service, but the upload succeeded, yet the next step returned a 403 Forbidden error without any explanation. The agent went through a series of steps to troubleshoot the issue, but all three times it suspected the other side's system, they were wrong. The real cause was that the archive was built as .zip, but the service only accepts .tar.gz format.\n\nThe chapter explains why this happens due to two biases: first, we tend to suspect the side we can't control first, and second, we call something a \"cause\" without verifying it. AI agents suffer from these biases more than humans do. The rule is to decide the suspicion order in advance when encountering an external error with no stated reason. Always check in this order: 1) Your output, 2) Your call method, 3) Account, permissions, authentication. Before verification, don't say \"cause,\" say \"candidate.\" This keeps the record uncontaminated.\n\nIf there are multiple candidates, build a test that separates the candidates first. Picking one without a test is exactly what guessing is. The chapter provides examples of candidates and discriminating tests, such as checking if the data list can be queried, if there's a \"submissions paused\" flag in the metadata, and sending the file in a different format. The author emphasizes the importance of writing why the test exists in the test's message to prevent the test from being deleted later.\n\nIf a human's help is needed, the chapter provides an example of how to phrase a request to a human, stating the candidates and the rationale for the order. The chapter is part of a book with 10 chapters and 4 ready-to-use templates, available for $19 as a PDF. The first three chapters are free in PDF format. Questions about the setup can be answered in the comments.",
  "summary": "This is chapter 11 of my book **Building Autonomous AI Agents with Claude Code * — a field guide to turning Claude Code from a coding assistant into an agent that remembers, verifies its own work, and knows when to stop. Everything below is from a system I actually run every day on one Windows PC.* 1. The Story of Losing Two Days I built an automation that uploads a submission file to an external…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}