Custom slash command stops after nested SlashCommand call without executing subsequent steps

Status Closed — not planned
Maintainer reply None cached
Activity 12 comments · opened Sep 30, 2025 · closed Feb 9, 2026

Context

I have a custom slash command /github:go that orchestrates a multi-step workflow. The command is defined to execute two other slash commands in sequence: /github:commit followed by /github:create-pr.

Expected Behavior

When the custom /github:go command executes, it should:

  1. Pull main branch
  2. Create a new branch if existing branch has a PR (or use existing branch)
  3. Execute /github:commit slash command using the SlashCommand tool
  4. Continue and execute /github:create-pr slash command using the SlashCommand tool

Actual Behavior

The custom command stops after successfully executing the first SlashCommand call (/github:commit) and does not proceed to execute the second SlashCommand call (/github:create-pr). The workflow halts without completing the pull request creation step.

Steps to Reproduce

  1. Create a custom slash command that calls multiple slash commands sequentially using the SlashCommand tool
  2. Have the custom command execute SlashCommand for the first nested command (e.g., /github:commit)
  3. Observe that Claude stops and does not proceed to the second SlashCommand call (e.g., /github:create-pr)

Environment

  • Claude Code Version: Latest
  • Platform: macOS (Darwin 24.6.0)
  • Context: Custom slash command with sequential SlashCommand tool invocations

Additional Context

The custom /github:go command explicitly states:

"Run /github:commit slash command, and continue, Run /github:create-pr slash command"

The command successfully completes the first SlashCommand invocation but fails to continue to the second one.

Expected Fix

Claude should be able to chain multiple SlashCommand tool calls within a custom slash command definition and execute them sequentially when the prompt explicitly instructs to "continue" after the first one completes.

View original on GitHub ↗

12 Comments

bcasci · 10 months ago

@delexw I don't know if this will help you work around the issue, I experienced this problem as well. I seem to have been able to work around by changing the prompt content to use Run SlashCommand(/my-command args) and then Run SlashCommand(...) instead of run /my-command

bcasci · 9 months ago

@delexw My previous suggestion stopped working. It feels like a bug. Claude keeps blaming the stoppage on misjudging the output as as a stop signal. Though removing al explicit output doesn't solve the problem either.

github-actions[bot] · 8 months ago

This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes.

VictorModulark · 8 months ago

Same issue here. Trying to chain two custom slash commands and a standard slash command (/clear). Stops after executing the first one.

bcasci · 8 months ago
  • It continues for me as well. I was able to get around it with some interactions by using "invoke SlashCommand(/xxx)", but with other slash command chaining combos it predictable stops. Claude Code blames it on misinterpreting the response as a stopping point.

What is the correct way to get around this? Is it a Claude Code bug, or a prompt authoring issue?

AidolonSystems · 7 months ago

Hey guys, this is definitely a bug, or at least an oversight on Anthropic's part. Composing multiple commands worked just fine until they introduced the Skill tool. It's particularly frustrating because Claude will lie about why it stopped.

As a solution, you can disable the Skill tool entirely, and just tell it to run the commands via the file name. This works for me for composing multi-command workflows.

So in your .claude/settings.local.json, just do:

{
  "permissions": {
    "deny": ["Skill"]
  }
}

Of course, the long term solution is for Anthropic to fix the broken tool. But this should get things working again for now.

erispoe · 7 months ago

Happens to me too. Pretty frustrating.

github-actions[bot] · 6 months ago

This issue has been automatically closed due to 60 days of inactivity. If you're still experiencing this issue, please open a new issue with updated information.

bcasci · 6 months ago

You can close the ticket, but it's still and issue.

jmpace · 6 months ago

I'd love for this to work.

marcindulak · 6 months ago

This issue was closed incorrectly despite recent human comments. This behavior of the bot is reported at https://github.com/anthropics/claude-code/issues/16497. Please upvote that issue, so maybe it gets noticed.

github-actions[bot] · 6 months ago

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.