Claude suggested a feature that does not work

Resolved 💬 3 comments Opened May 1, 2026 by maaspet Closed May 5, 2026

Preflight Checklist

  • [x] I have searched existing issues and this hasn't been reported yet
  • [x] This is a single bug report (please file separate reports for different bugs)
  • [x] I am using the latest version of Claude Code

What's Wrong?

  • Account UUID: 6aa0f05d-998f-46b2-97a5-975a86c2ac08
  • Environment: env_01L1FoxZh3N96umAbYo5G2tJ (just-created Default env)
  • Two routines failing to start: trig_01FLV74mmerPErGSwg48Z7mT (smoke test, public repo, sonnet) + trig_01HHQndCfUyydBg9WApSPoTA (rangeAccrual, your real one)
  • Symptoms: UI "Failed to start run" with no detail; scheduled run_once_at passes with ended_reason staying empty.

What Should Happen?

Agent should start

Error Messages/Logs

Steps to Reproduce

  1. From a brand-new account with no pre-existing CCR environment, invoke the /schedule skill in Claude Code. The skill auto-creates a Default environment.
  2. Create a routine with a single source pointing at any public GitHub repo, a trivial prompt (e.g. Run "ls /workspace" via Bash and report.), enabled: true, and run_once_at set ~3 minutes in the future.
  3. Wait until past run_once_at.

Claude Model

Not sure / Multiple models

Is this a regression?

No, this never worked

Last Working Version

_No response_

Claude Code Version

2.1.126 (Claude Code)

Platform

Anthropic API

Operating System

Ubuntu/Debian Linux

Terminal/Shell

Other

Additional Information

_No response_

View original on GitHub ↗

This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗