claude-code - 💡(How to fix) Fix Feature Request: Separate /-prefix (commands) from skills/agents picker (Codex-style) [1 comments, 2 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#58787Fetched 2026-05-14 03:39:30
View on GitHub
Comments
1
Participants
2
Timeline
5
Reactions
0
Author
Timeline (top)
labeled ×4commented ×1

Code Example

//help, /clear, /config, /loop, /review, /plugin,$        → memsearch:memory-recall, coderabbit:code-review, superpowers:*,
RAW_BUFFERClick to expand / collapse

Motivation

Currently, the / prefix in Claude Code shows a mixed list of:

  • Built-in slash commands (/help, /clear, /config, /loop, …)
  • User-defined slash commands
  • Skills (auto-installed + plugin-provided)
  • Agents / subagents

With many plugins/skills installed (superpowers, pr-review-toolkit, coderabbit, memsearch, frontend-design, skill-creator, …) the / menu becomes very long and hard to scan. Finding a built-in command means scrolling past dozens of skills.

The wording in claude --help reinforces that skills and slash commands share the same pipeline:

  • --disable-slash-commands is documented as "Disable all skills"
  • --bare says "Skills still resolve via /skill-name"

That is a deliberate design choice — but as the skill ecosystem grows it becomes a discoverability problem.

Proposed Solution

Adopt a prefix separation similar to OpenAI Codex CLI:

  • / → only built-in + user-defined slash commands
  • $ (or another prefix like @@) → Skills and/or agent picker (own namespace)

The @ prefix is already taken in Claude Code (file references), so $ is the obvious free key.

Benefits

  • Clean, predictable / menu — easy to find built-ins again
  • Faster category-based discovery
  • Clearer mental model: commands vs extensions
  • Plugin authors still get a first-class entry point (just under a different prefix)

Alternatives considered

  • skillOverrides: "off" per skill — works, but is opt-out per skill and removes them entirely (no picker entry at all). Doesn't solve discoverability; it solves clutter only by deletion.
  • Disabling whole plugins via enabledPlugins — same issue, all-or-nothing.
  • A simple toggle / settings flag (e.g. skillsInSlashPicker: false) would already be enough — doesn't have to be a new prefix.

Example

/        → /help, /clear, /config, /loop, /review, /plugin, …
$        → memsearch:memory-recall, coderabbit:code-review, superpowers:*, …

Environment

  • Claude Code CLI
  • macOS 15.5.0
  • Many plugins installed via the official marketplace

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 Feature Request: Separate /-prefix (commands) from skills/agents picker (Codex-style) [1 comments, 2 participants]