Skip to content

Extract agent guidance into skills docs#969

Merged
0xh3rman merged 2 commits intomainfrom
agent-skills
Feb 26, 2026
Merged

Extract agent guidance into skills docs#969
0xh3rman merged 2 commits intomainfrom
agent-skills

Conversation

@0xh3rman
Copy link
Collaborator

  • Split the large AGENTS.md into focused skill documents and update the top-level guidance.
  • AGENTS.md now points to mandatory reading under skills/ and includes a short formatting note; the long inline content was moved into new files:
    • skills/architecture.md, skills/code-style.md
    • skills/common-issues.md
    • skills/defensive-programming.md
    • skills/development-commands.md
    • skills/error-handling.md
    • skills/project-structure.md
    • skills/tests.md.
  • Also condensed the task completion checklist in AGENTS.md.

Split the large AGENTS.md into focused skill documents and update the top-level guidance. AGENTS.md now points to mandatory reading under skills/ and includes a short formatting note; the long inline content was moved into new files: skills/architecture.md, skills/code-style.md, skills/common-issues.md, skills/defensive-programming.md, skills/development-commands.md, skills/error-handling.md, skills/project-structure.md, and skills/tests.md. Also condensed the task completion checklist in AGENTS.md.
@semanticdiff-com
Copy link

Review changes with  SemanticDiff

@gemini-code-assist
Copy link
Contributor

Summary of Changes

Hello @0xh3rman, 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 the agent guidance documentation by modularizing a large, comprehensive Markdown file into a collection of specialized skill documents. This change aims to enhance the clarity and accessibility of development guidelines, making it easier for AI assistants to navigate and understand specific aspects of the codebase, from project architecture to coding standards and testing practices.

Highlights

  • Documentation Restructuring: The monolithic AGENTS.md file has been split into several smaller, focused skill-specific Markdown documents to improve readability and maintainability for AI assistants.
  • New Skill Documents: Eight new documents were created under a skills/ directory: architecture.md, code-style.md, common-issues.md, defensive-programming.md, development-commands.md, error-handling.md, project-structure.md, and tests.md.
  • Updated Top-Level Guidance: AGENTS.md now serves as a high-level entry point, directing users to the mandatory skill documents and including a concise formatting note.
  • Condensed Task Completion Checklist: The task completion checklist within AGENTS.md was streamlined for brevity.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog
  • AGENTS.md
    • Removed extensive sections covering project overview, repository layout, technology stack, development workflow, coding standards, architecture, and testing.
    • Added a 'Skills' section with links to eight new, dedicated skill documents.
    • Introduced a 'Formatting (mandatory)' section with specific rustfmt and cargo clippy commands.
    • Condensed the 'Task Completion' checklist.
  • skills/architecture.md
    • Added new file detailing key architectural principles, provider/mapper patterns, repository patterns, RPC client patterns, UniFFI patterns, and shared utilities.
  • skills/code-style.md
    • Added new file outlining code formatting rules, commit message guidelines, naming conventions, import ordering, and general code organization principles.
  • skills/common-issues.md
    • Added new file documenting known anti-patterns and their recommended fixes, including alloy_primitives::hex usage, a misspelled file, duplicate constants, inline Diesel imports, println! usage, and technical debt markers.
  • skills/defensive-programming.md
    • Added new file specifying safety rules such as avoiding matches!, #[allow(dead_code)], todo!()/unimplemented!(), println!, .unwrap()/.expect(), and promoting immutable bindings.
  • skills/development-commands.md
    • Added new file listing just commands for various development tasks including building, testing, code quality checks, database operations, mobile setup, and binding generation.
  • skills/error-handling.md
    • Added new file providing guidance on error type patterns, constructor methods, From trait implementations for the ? operator, JSON parameter extraction, database patterns, and async patterns.
  • skills/project-structure.md
    • Added new file describing the overall project structure, including applications, the cross-platform library, blockchain support, utility binaries, core crates, and the technology stack.
  • skills/tests.md
    • Added new file detailing testing conventions, appropriate use of .unwrap() in tests, test data management, mock patterns with testkit/ modules, direct assert_eq!, test helpers, and integration testing guidelines.
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.

Copy link
Contributor

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request is a great improvement, refactoring the large AGENTS.md file into smaller, more focused 'skill' documents. This makes the guidance for AI assistants much more modular and easier to navigate. The new structure is clear and the content is well-organized. I have a couple of minor suggestions to further improve the new documentation by removing some redundancy and clarifying a heading.

@0xh3rman 0xh3rman marked this pull request as ready for review February 26, 2026 02:50
@0xh3rman 0xh3rman merged commit 8829fc6 into main Feb 26, 2026
4 checks passed
@0xh3rman 0xh3rman deleted the agent-skills branch February 26, 2026 02:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant