Uxc Skill Creator

Create wrapper skills that call remote tools through UXC. Use when defining a new provider skill and you need reusable templates, validation rules, and anti-pattern guidance based on proven UXC skill practices.

Use this skill to design and standardize provider wrapper skills built on top of uxc.

Prerequisites

• uxc skill is available as the base execution contract.
• Target wrapper skill scope is clear (provider endpoint, core operations, auth model, write risk).

Output Contract

A wrapper skill created with this skill should include:

• SKILL.md
• agents/openai.yaml
• references/usage-patterns.md
• scripts/validate.sh

Optional files are allowed only when they add real reusable value.

Core Workflow

1. Start from user-provided host input:
   • record the raw host the user gives
   • normalize endpoint candidates (scheme/no-scheme, path variants)
2. Discover protocol and valid path before drafting skill text:
   • search official docs/repo to confirm endpoint shape and auth model
   • probe candidates with uxc <endpoint> -h
   • confirm one working endpoint + protocol as the wrapper target
3. Detect authentication requirement explicitly:
   • run host help or a minimal read call and inspect envelope/error code
   • if auth-protected, record required model (api key or oauth) and scopes
   • verify local mapping path with uxc auth binding match <endpoint> when OAuth/binding is used
4. Fix the wrapper interface:
   • provider endpoint (<host>)
   • fixed link command name (<provider>-<protocol>-cli)
   • auth mode (none, api key, oauth)
5. Write SKILL.md as a thin execution policy:
   • link-first command flow
   • help-first discovery flow
   • JSON envelope parsing and safe-write guardrails
6. Add provider-specific references/usage-patterns.md:
   • minimal read and write examples
   • key=value and bare JSON positional input examples
7. Add scripts/validate.sh with strict checks:
   • required files
   • frontmatter fields
   • command style constraints
   • banned legacy patterns
8. Add agents/openai.yaml for skill UI metadata.
9. Run validation and iterate until clean.

Hard Rules

• Default to link-first (command -v <link_name> then uxc link <link_name> <host>).
• Default to help-first (<link_name> -h, <link_name> <operation> -h).
• Use protocol-aware link naming:
  • format: <provider>-<protocol>-cli
  • examples: notion-mcp-cli, github-openapi-cli
• Prefer key=value; allow bare JSON positional payload.
• Keep JSON output as automation path; do not rely on --text.
• Do not use legacy default examples (list/describe/call/removed flags).
• Do not use dynamic link renaming at runtime.
• Do not assume protocol/path/auth from host string alone; verify by search + probe.

References

• Step-by-step implementation flow:
  • references/workflow.md
• Copy-ready templates:
  • references/templates.md
• Validation checklist and banned patterns:
  • references/validation-rules.md
• Observed pitfalls and better defaults:
  • references/anti-patterns.md

See Also

• Base execution and protocol/auth guidance:
  • skills/uxc/SKILL.md