claude-code - 💡(How to fix) Fix Show provenance/origin of slash commands (native, plugin, skill, author) [1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#48708Fetched 2026-04-16 06:53:12
View on GitHub
Comments
0
Participants
1
Timeline
4
Reactions
1
Participants
Timeline (top)
labeled ×4

When browsing available slash commands (via / autocomplete or /help), there's no way to tell where a command comes from or who created it. As the ecosystem of plugins and skills grows, this becomes a real gap — both for discoverability and security/auditing.

Root Cause

  • Security/auditing: In team environments, knowing whether a command was added by a trusted plugin vs. an unknown skill is important. Skills execute with full agent capabilities.
  • Discoverability: When you have 50+ commands, knowing which are native vs. plugin-provided helps users understand what's available and why.
  • Debugging: When a slash command behaves unexpectedly, knowing its source is the first step to investigating.
  • Trust signals: As the skill/plugin ecosystem grows, users need to know what they're running — similar to how VS Code shows extension publishers.

Fix Action

Fix / Workaround

Current Workarounds

Code Example

---
   name: my-skill
   description: Does a thing
   author: "Anthropic"
   source: "plugin:vercel"
   ---
RAW_BUFFERClick to expand / collapse

Summary

When browsing available slash commands (via / autocomplete or /help), there's no way to tell where a command comes from or who created it. As the ecosystem of plugins and skills grows, this becomes a real gap — both for discoverability and security/auditing.

Problem

A typical session might have 50+ slash commands from mixed sources:

  • Native CLI commands built by Anthropic (/help, /config, /compact)
  • Plugin skills from third parties (vercel:deploy, gstack:browse)
  • Project-local skills from .claude/skills/
  • User-global skills from ~/.claude/skills/

Today, the only way to tell them apart is by naming convention (namespaced prefixes) or by manually inspecting SKILL.md files. There's no metadata in the UI, no author field in skill frontmatter, and no visual distinction in the command list.

Proposed Solution

  1. Show provenance in the / autocomplete and /help output — a tag or grouping like [native], [plugin: vercel], [skill: project], [skill: user]
  2. Add optional author and source fields to skill frontmatter so skill creators can declare who made them:
    ---
    name: my-skill
    description: Does a thing
    author: "Anthropic"
    source: "plugin:vercel"
    ---
  3. Group or filter commands by source — e.g., /help --source plugin or visual grouping in the autocomplete menu

Why This Matters

  • Security/auditing: In team environments, knowing whether a command was added by a trusted plugin vs. an unknown skill is important. Skills execute with full agent capabilities.
  • Discoverability: When you have 50+ commands, knowing which are native vs. plugin-provided helps users understand what's available and why.
  • Debugging: When a slash command behaves unexpectedly, knowing its source is the first step to investigating.
  • Trust signals: As the skill/plugin ecosystem grows, users need to know what they're running — similar to how VS Code shows extension publishers.

Current Workarounds

  • Namespaced commands (e.g., vercel:deploy) hint at the source, but this is convention, not enforced
  • Manually reading SKILL.md files or checking .claude/plugins/ directories
  • Memorizing which commands are built-in vs. third-party

🤖 Generated with Claude Code

extent analysis

TL;DR

To address the issue of identifying the source of slash commands, add provenance information to the / autocomplete and /help output, and introduce optional author and source fields to skill frontmatter.

Guidance

  • Introduce a tagging system in the / autocomplete and /help output to indicate the source of each command, such as [native], [plugin: vercel], [skill: project], or [skill: user].
  • Add optional author and source fields to skill frontmatter to allow skill creators to declare their identity and the source of the skill.
  • Consider grouping or filtering commands by source in the /help output or autocomplete menu to improve discoverability and security.
  • Update the UI to display the provenance information and provide visual distinctions between commands from different sources.

Example

---
name: my-skill
description: Does a thing
author: "Anthropic"
source: "plugin:vercel"
---

Notes

The proposed solution requires updates to the UI and the skill frontmatter format. It is essential to ensure backward compatibility and provide clear documentation for skill creators to use the new fields.

Recommendation

Apply the proposed solution to add provenance information and introduce optional author and source fields to skill frontmatter, as it addresses the security, discoverability, and debugging concerns.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

claude-code - 💡(How to fix) Fix Show provenance/origin of slash commands (native, plugin, skill, author) [1 participants]