Skip to content

Add generic tmt context to AGENTS.md#4585

Open
happz wants to merge 5 commits into
mainfrom
extend-claude-context
Open

Add generic tmt context to AGENTS.md#4585
happz wants to merge 5 commits into
mainfrom
extend-claude-context

Conversation

@happz
Copy link
Copy Markdown
Contributor

@happz happz commented Feb 16, 2026

And AGENTS.md is born from Claude instructions. This way we should cover other agents as well as Claude.

Pull Request Checklist

  • implement the feature

@happz happz added this to planning Feb 16, 2026
@happz happz added the code | no functional change "No Functional Change" intended. Patch should not change tmt's behavior in any way. label Feb 16, 2026
@github-project-automation github-project-automation Bot moved this to backlog in planning Feb 16, 2026
@happz happz moved this from backlog to review in planning Feb 16, 2026
@gemini-code-assist
Copy link
Copy Markdown
Contributor

Summary of Changes

Hello @happz, 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 enhances the .claude/CLAUDE.md file by incorporating a comprehensive guide to the TMT project. The aim is to provide AI assistants with detailed context about the codebase, enabling them to offer more accurate and relevant assistance to developers working on TMT.

Highlights

  • AI Assistant Context: Added extensive documentation regarding the TMT (Test Management Tool) project, including its overview, development commands, architecture, configuration, testing structure, CLI, and coding standards, to the .claude/CLAUDE.md file.
Changelog
  • .claude/CLAUDE.md
    • Added a detailed overview of the TMT project, including its architecture, development workflows, and coding standards, to provide better context for AI assistants.
Activity
  • No specific activity or review comments were recorded for this pull request.
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
Copy Markdown
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 adds a comprehensive guide for AI assistants to the .claude/CLAUDE.md file. The guide provides a good overview of the tmt project, including its architecture, development commands, and coding standards. The information is largely accurate and will be very helpful. I've suggested one minor improvement to the 'Development Commands' section to correct the command for running tests and remove a redundancy.

Comment thread .claude/CLAUDE.md Outdated
@happz happz force-pushed the extend-claude-context branch from 123232e to 12fcda1 Compare March 2, 2026 07:53
@vaibhavdaren vaibhavdaren self-requested a review May 6, 2026 11:50
@psss psss added this to the 1.74 milestone May 7, 2026
@happz happz force-pushed the extend-claude-context branch from 12fcda1 to ec08df9 Compare May 7, 2026 19:13
@happz happz changed the title Add generic tmt context to Claude instructions Add generic tmt context to AGENTS.md May 7, 2026
Comment thread AGENTS.md Outdated
Comment thread AGENTS.md Outdated
Comment thread AGENTS.md Outdated

- Unit tests should mock external dependencies
- Integration tests can use the container framework
- Follow existing test patterns in `tests/` directory
Copy link
Copy Markdown
Contributor

@vaibhavdaren vaibhavdaren May 11, 2026

Choose a reason for hiding this comment

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

Too ambiguous for an AI. You are forcing the agent to guess which file represents your gold standard.
Give it a specific anchor. Change this to: "Follow existing test patterns (e.g., refer to tests/prepare/install)"

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Which test shall I recommend? @psss @thrix @LecrisUT any tips?

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

/tests/prepare/install probably because it has the context of running on multiple images.

Copy link
Copy Markdown
Contributor Author

@happz happz May 18, 2026

Choose a reason for hiding this comment

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

Added in 0b73a5f.

Comment thread AGENTS.md Outdated
Comment thread AGENTS.md Outdated
- Security-focused linting with bandit checks provided by ruff


## Release Notes Generation
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

if you load this entire AGENTS.md file into your AI's context every time you start a chat, you are forcing the AI to read and "remember" how to format release notes even when you just want it to help you debug a unit test in tmt/steps/provision. So this should be moved to a seprate file.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Which file then? If these are instructions for an agent which shall be in the AGENTS.md file, where should we put instructions that are supposed to be read by an agent, but only in some situations?

Copy link
Copy Markdown
Contributor

@vaibhavdaren vaibhavdaren May 14, 2026

Choose a reason for hiding this comment

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

I'd suggest moving the actual release note instructions into a dedicated file, something like docs/agents/release-notes-prompt.md or somewhere else.

What you keep in AGENTS.md is just a pointer to that new file. You could add a simple rule in AGENTS.md that says something like:

If the task involves generating release notes,
load and follow the instructions in docs/agents/release-notes-prompt.md.

This way, AGENTS.md stays lean and acts as a router. The heavy, specific formatting instructions are only loaded into the AI's context when release note generation is explicitly requested. If a developer is just asking the AI to debug a unit test, those extra instructions are ignored, keeping the AI focused entirely on the code.

@thrix is there anything better we could do here?

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

something like docs/agents/release-notes-prompt.md or somewhere else.

Can we please not contaminate other directories with machine language

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

What do you suggest then as an alternative? Keeping everything related to agents in a single huge file? That does not seem to be very maintainable to me. Having a dedicated folder for agent skills where each area would have its only file sounds reasonable to me. If we agree on the specific location there will not be any "contamination" with other docs.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

What do you suggest then as an alternative? Keeping everything related to agents in a single huge file? That does not seem to be very maintainable to me.

The thing is not human readable/maintainable from the beginning IMO.

Having a dedicated folder for agent skills where each area would have its only file sounds reasonable to me.

Yes, a dedicated folder would be preferred.

Copy link
Copy Markdown
Contributor Author

@happz happz May 18, 2026

Choose a reason for hiding this comment

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

Yes, let's keep assistant instructions separated from user-focused documentation. Moved in 0b73a5f.

@psss psss self-assigned this May 12, 2026
@happz happz force-pushed the extend-claude-context branch from ec08df9 to cd34663 Compare May 14, 2026 07:30
@psss psss requested a review from vaibhavdaren May 14, 2026 13:54
Comment thread AGENTS.md Outdated

7. **Review**: Ensure all significant changes for the milestone are covered and the formatting matches the existing style in the `docs/releases/` directory.

## Using This Instruction with Different AI Tools
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

This section and the following ones (MCP Setup and Custom Instructions) are human-oriented documentation, not an interesting reading for agents, let's move them to the contribute page.

Copy link
Copy Markdown
Contributor Author

@happz happz May 18, 2026

Choose a reason for hiding this comment

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

Moved in 0b73a5f.

Comment thread AGENTS.md Outdated

## Overview

TMT (Test Management Tool) is a comprehensive Python-based testing framework that provides
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Suggested change
TMT (Test Management Tool) is a comprehensive Python-based testing framework that provides
tmt (Test Management Tool) is a comprehensive Python-based testing framework that provides

tmt and fmf should be always written in lower-case, except for envvar, class and similar exceptions...

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Fixed in 0b73a5f.

@happz happz self-assigned this May 17, 2026
@happz happz force-pushed the extend-claude-context branch 2 times, most recently from 1754050 to 6e44ff1 Compare May 18, 2026 07:07
@happz happz force-pushed the extend-claude-context branch from 6e44ff1 to 0b73a5f Compare May 18, 2026 07:10
@happz happz removed their assignment May 18, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

code | no functional change "No Functional Change" intended. Patch should not change tmt's behavior in any way.

Projects

Status: review

Development

Successfully merging this pull request may close these issues.

6 participants