claude-code - 💡(How to fix) Fix [BUG] the connector-query API the skill uses is returning empty when connectors are present.

Preflight Checklist

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

What's Wrong?

What Should Happen?

Two real issues. Let me address both directly.

Issue 1 — skill template fix: You're right, I named the root cause but didn't fix it. Let me look at whether there's a local skill file I can patch, and check current permission settings at the same time.

Found files, read 2 files Now I have the full picture. Let me be precise about both issues:

Issue 1 — Skill template: The schedule skill is Anthropic-hosted. I cannot edit its source. The "No connected MCP connectors" text is dynamically generated by a connector-query API that returns empty even when connectors exist — that's a bug on Anthropic's side. What I can do: inject the fix via a SessionStart hook so it's in context before the skill ever runs. I'll do that below.

Issue 2 — Permission prompts: The project settings.json is missing allowlist entries for every tool used in the schedule workflow. I'll add them now.

Edited settings.json +17 -1 Now fix the SessionStart hook to inject the known Slack connector facts so the skill template lie is corrected before any skill reads it:

Read a file, edited a file Done. Here's what was actually fixed and why each fix targets the root cause:

Error Messages/Logs

Steps to Reproduce

Two real issues. Let me address both directly.

Issue 1 — skill template fix: You're right, I named the root cause but didn't fix it. Let me look at whether there's a local skill file I can patch, and check current permission settings at the same time.

Found files, read 2 files Now I have the full picture. Let me be precise about both issues:

Issue 1 — Skill template: The schedule skill is Anthropic-hosted. I cannot edit its source. The "No connected MCP connectors" text is dynamically generated by a connector-query API that returns empty even when connectors exist — that's a bug on Anthropic's side. What I can do: inject the fix via a SessionStart hook so it's in context before the skill ever runs. I'll do that below.

Issue 2 — Permission prompts: The project settings.json is missing allowlist entries for every tool used in the schedule workflow. I'll add them now.

Edited settings.json +17 -1 Now fix the SessionStart hook to inject the known Slack connector facts so the skill template lie is corrected before any skill reads it:

Read a file, edited a file Done. Here's what was actually fixed and why each fix targets the root cause:

Claude Model

Sonnet (default)

Is this a regression?

I don't know

Last Working Version

No response

Claude Code Version

Claude Code 2.1.150

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

Cursor

Additional Information

No response