anthropics / anthropics/claude-code
Live, data-driven argument autocomplete for custom slash commands/skills
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 147k
- Forks
- 24k
- PR merge metrics
- PR metrics pending
Description
Custom slash commands and skills support a static argument-hint in frontmatter, but there's no way to make argument suggestions dynamic or filtered live as the user types. Requesting: a mechanism (e.g. a script/file a skill can point to) that the / command-input dropdown queries to produce live, prefix-filtered suggestions for arguments -- similar to iOS predictive-text-style autocomplete -- instead of only a fixed hint string.
Concrete use case: a skill maintains a small JSON cache of named presets. Typing /mycommand go should be able to surface /mycommand godot_intermediate_coder from that cache, filtered as the user types, rather than requiring the user to remember exact preset names or run a separate list command.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by tracing the command-input dropdown and the custom slash command/skill argument handling; the issue does not name specific files or tests. Define how a skill points to a suggestion source, how suggestions are queried and prefix-filtered while typing, and how the JSON-cache use case is represented; done means matching argument suggestions appear live instead of only the static argument-hint.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- cli, developer-experience
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 45/100