skill-prompt-convert

Original🇨🇳 Chinese
Translated

Convert between Skill (SKILL.md) and Prompt (chat box instruction) formats. Supports two-way conversion between Skill→Prompt and Prompt→Skill with zero loss of core information. Use this when users mention Skill to Prompt conversion, Prompt to Skill conversion, format mutual conversion, or SKILL.md conversion.

3installs
Added on

NPX Install

npx skill4agent add xstongxue/best-skills skill-prompt-convert

SKILL.md Content (Chinese)

View Translation Comparison →

Mutual Conversion between Skill and Prompt

Usage Scenarios

  • Users need to convert Skill to a Prompt that can be directly copied to the chat box
  • Users need to convert Prompt to Skill in SKILL.md format
  • Users mention mutual conversion between Skill and Prompt, or format conversion

Step 1: Obtain Input and Confirm Conversion Direction

Obtain the content to be converted from the user's message, paste, or attachment, and determine the conversion direction:
  • A. Skill → Prompt: Input is SKILL.md content, output is a Prompt usable in the chat box
  • B. Prompt → Skill: Input is Prompt content (including ``` code blocks), output is in SKILL.md format

Step 2: Execute Conversion

When converting Skill → Prompt: Integrate
name
and
description
into
# Role
and
# Task
; Convert "Usage Scenarios" to applicable scenarios; Convert Step 1/2... into
# Workflow
or
# Constraints
; Add
# Output Format
and
# Input
placeholders; Wrap the output with ```.
When converting Prompt → Skill: Extract
name
(lowercase English with hyphens) and
description
from
# Role
and
# Task
; Extract "Usage Scenarios" from
# Task
or
# Constraints
; Break down
# Workflow
and
# Constraints
into Step 1, Step 2...; Retain
# Output Format
in "Notes"; Output the complete SKILL.md, which must comply with the standard Skill format.

Step 3: Output Results

  • Part 1 [Conversion Explanation]: Briefly explain the mapping relationship (e.g., Step 1 is mapped to the first step of Workflow)
  • Part 2 [Conversion Result]: Complete converted content (can be directly pasted for use)

Notes

  • Zero Information Loss: Do not delete core logic, steps, or constraints
  • Format Adaptation: Skill emphasizes "when to trigger" and "how the Agent executes"; Prompt emphasizes "what the user inputs" and "what output format to use"
  • Naming Convention:
    name
    must be in lowercase English with hyphens (e.g.,
    wechat-article-writer
    ); reference file names follow the same rule (e.g.,
    outline-review-science.md
    )
  • Description Quality: Must include "when to use", "what it's used for", and key words
  • Skill Body Structure: Title → Usage Scenarios → Step 1/2... → Notes/Examples; For multi-input tasks, supplement "input method when triggered"
  • Standard Skill Format: YAML frontmatter must start and end with
    ---
    , using
    name:
    and
    description:
    (strictly prohibit
    ## name:
    ), example:
    ---
    name: skill-name
    description: Trigger scenarios and function description
    ---
    
    # Title
    ...