Coze Skill Creator

Task Objectives

• This Skill is used to: Create complete skills from configuration files or requirement descriptions, including tool configuration, workflow orchestration, and code generation
• Capabilities include: Configuration validation, file generation, template rendering, skill packaging
• Trigger conditions: Users describe skill requirements, provide skill configuration JSON, design workflow structures

Prerequisites

• Dependency description: Dependencies and versions required by scripts

  jsonschema>=4.0.0

• Non-standard files/folders preparation: None

Operation Steps

• Standard workflow:
  1. Requirement Analysis
     • The agent understands the core capabilities, trigger scenarios, and tool requirements of the skill based on the user's description
     • Refer to references/config-reference.md to determine the configuration structure
  2. Configuration Generation
     • Generate a skill configuration JSON file based on the analysis results
     • Configuration includes: Metadata, tool list, workflow nodes, script definitions
  3. Schema Validation
     • Call

       scripts/validate_schema.py --config <config_path>

       to verify the configuration format
     • If validation fails, correct the configuration based on error messages
  4. File Generation
     • Call

       scripts/generate_skill.py --config <config_path> --output <output_dir>

       to generate skill files
     • The script automatically creates the structure of SKILL.md, scripts/, references/, assets/
  5. Packaging and Testing
     • Use the

       package_skill(skill_dir_name=<skill_name>)

       tool to package the skill
     • Verify that the .skill file is generated successfully and has complete content
• Optional branches:
  • When the user provides configuration JSON: Skip directly to step 3 for validation
  • When only workflow design is needed: Refer to references/workflow-guide.md, the agent generates workflow diagrams and node definitions
  • When only tool integration configuration is needed: Refer to references/tool-reference.md, the agent generates tool configurations

Resource Index

• Required scripts:
  • scripts/validate_schema.py (Purpose and parameters: Verify whether the skill configuration JSON conforms to the Schema, parameter: --config)
  • scripts/generate_skill.py (Purpose and parameters: Generate complete skill files based on configuration, parameters: --config, --output)
• Domain references:
  • references/config-reference.md (When to read: When generating or validating configurations)
  • references/workflow-guide.md (When to read: When designing workflows)
  • references/tool-reference.md (When to read: When configuring tools)
• Output assets:
  • assets/templates/skill-template.md (Directly used to generate SKILL.md)
  • assets/templates/python-script.py (Directly used to generate Python scripts)
  • assets/templates/schema.json (Used to verify configuration formats)

Notes

• The configuration JSON must conform to the Schema defined in assets/templates/schema.json
• Generated skill names must use lowercase letters and hyphens, and the -skill suffix is prohibited
• The connection relationship of workflow nodes must maintain a directed acyclic graph structure
• Make full use of the agent's natural language understanding ability, and only call scripts when structured output is required

Usage Examples

Example 1: Create a skill from a requirement description

• Function description: User describes "Need a skill to handle image compression"
• Execution method: Agent analyzes requirements → generates configuration → script validation → generates files → packaging
• Key parameters: Skill name image-compressor, description, tool list
• Configuration example:
  json

  {
    "name": "image-compressor",
    "description": "Compress image file size",
    "tools": [
      {
        "name": "compress",
        "type": "python",
        "parameters": [{"name": "quality", "type": "integer"}]
      }
    ]
  }

Example 2: Generate a skill from a configuration file

• Function description: The user has provided a skill-config.json configuration file
• Execution method: Script validation → script generation → packaging
• Key parameters: --config skill-config.json, --output ./my-skill
• Command:

  python scripts/generate_skill.py --config skill-config.json --output ./my-skill

Example 3: Design a workflow

• Function description: The user needs to create a skill workflow with branch logic
• Execution method: Agent reads workflow guide → designs nodes → generates configuration → generates skill
• Key parameters: Node type, connection relationship, conditional expression
• Workflow example:

  [Start] → [Condition Judgment] → [Branch A: API Call] → [End]
                    ↘ [Branch B: Local Processing] → [End]