feat(skill): update skill generation logic

[helloeve]

This PR refactors and improves the skill generation logic to make it more configurable and also simplifies the generated skill.

• Shift from generating individual tool-specific YAML files to a more centralized configuration approach using global toolbox flags.
• Add support for --license-header to prepend license information to generated Node.js scripts.
• Refactored tool parameter documentation from a JSON schema format to a more readable Markdown table.
• Update tool invocation log level to avoid unnecessary outputs

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request significantly refactors and enhances the skill generation logic within the toolbox. The primary goal is to improve configurability and simplify the structure of generated skills. This is achieved by centralizing tool configuration, allowing for dynamic passing of settings, and introducing the ability to include license headers in generated scripts. Additionally, the readability of skill documentation is improved by presenting tool parameters in a more user-friendly Markdown table format, and the verbosity of tool execution logs is reduced.

Highlights

• Centralized Skill Configuration: Shifted from generating individual tool-specific YAML files to a more centralized configuration approach, passing tool configurations directly as arguments to the Node.js wrapper scripts.
• License Header Support: Added a new --license-header flag to prepend license information to generated Node.js scripts, improving compliance and code attribution.
• Improved Parameter Documentation: Refactored tool parameter documentation in SKILL.md files from a JSON schema format to a more readable Markdown table.
• Reduced Log Output: Updated the tool invocation log level to 'error' to avoid unnecessary outputs during skill execution, making logs cleaner.

Changelog

• cmd/internal/skills/command.go
  • Imported the strings package.
  • Added a licenseHeader field to the skillsCmd struct.
  • Introduced a new --license-header flag for the skills command.
  • Implemented logic to dynamically construct jsConfigArgs based on prebuilt configurations, tool folders, and tool files, copying necessary assets to the skill directory.
  • Removed the generation of individual tool-specific YAML configuration files.
  • Updated the generateScriptContent function call to pass the new configArgsStr and cmd.licenseHeader.
  • Added copyFile and copyDir helper functions for asset management.
• cmd/internal/skills/command_test.go
  • Updated the expected asset path in tests to reflect the removal of individual tool YAML files, now expecting a generic tools.yaml.
• cmd/internal/skills/generator.go
  • Removed unused imports: bytes, encoding/json, goccy/go-yaml, internal/server, internal/sources.
  • Modified the nodeScriptTemplate to conditionally include a license header and removed the toolsFileName variable.
  • Updated the markdown template for Bash and PowerShell commands to include <skill_dir> in the script path.
  • Adjusted the generateSkillMarkdown comment to reflect the change in parameter documentation.
  • Modified the nodeScriptTemplate to use a ConfigArgs variable directly and added --log-level error to the toolboxArgs.
  • Updated the scriptData struct to include ConfigArgs and LicenseHeader, and removed ToolsFileName.
  • Refactored the generateScriptContent function signature and implementation to accept configArgs and licenseHeader.
  • Rewrote the formatParameters function to generate a Markdown table for parameters instead of a JSON schema.
  • Removed the generateToolConfigYAML function, which was responsible for generating individual tool YAML files.
• cmd/internal/skills/generator_test.go
  • Updated TestFormatParameters assertions to expect Markdown table output instead of JSON schema.
  • Modified TestGenerateSkillMarkdown assertions to reflect the updated script path and Markdown table format for parameters.
  • Updated TestGenerateScriptContent tests to cover the new configArgs and licenseHeader parameters, and removed tests related to toolsFileName.
  • Removed the TestGenerateToolConfigYAML function and its associated tests.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request significantly improves the skill generation logic by shifting from individual tool-specific YAML files to a centralized configuration and enhancing documentation readability by rendering tool parameters as Markdown tables. However, it introduces several code injection vulnerabilities where user-controlled inputs from command-line flags and filesystem names are injected into generated Node.js wrapper scripts without proper escaping or sanitization, potentially allowing arbitrary code execution. Additionally, the generated Markdown documentation can be malformed if a parameter description contains a pipe character. Addressing these security and documentation issues is crucial for robustness.

[github-actions]

🔎 Preview at https://googleapis.github.io/genai-toolbox/previews/PR-2646/

[github-actions]

🧨 Preview deployments removed.