编程

UXC Skill Creator

试用

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-...

它能做什么

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-...

技能文档

UXC Skill Creator

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 -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 when OAuth/binding is used
  4. Fix the wrapper interface:
    • provider endpoint (``)
    • fixed link command name (--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 then uxc link ).
  • Default to help-first ( -h, -h).
  • Use protocol-aware link naming:
    • format: --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

相关技能

Discover and call remote schema-exposed interfaces with UXC. Use when an agent or skill needs to list operations, inspect operation schemas, and execute Open...

36 次安装

Build a NEW agent skill from scratch, end-to-end — a thin conductor dispatches fresh subagents through five gated roles (compose spec -> design structure -> red-green build -> compress -> independent attack). EXPENSIVE (large token cost): trigger ONLY on an explicit user request to author/build/create an agent skill — "build me a skill", "create a new skill", "package this repeated workflow so it triggers automatically", "$skill-creator-max". Do-NOT fire for: summarizing or writing daily/session memory or journaling (incl. Chinese "总结/记录今天的记忆"), or any generic "create/make/summarize X" that is not authoring an agent skill.

Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.

3 次安装1 星标

End-to-end guided workflow for building production-quality agent skills. Walks users through 4 stages: pre-flight checks → spec confirmation (3 musts) → codi...

Discover, install, update, or create the right skill when a workflow gap appears. Use when a task is repetitive, a role lacks a reliable procedure, an existi...

37 次安装1 星标

Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.

25 次安装1 星标