microsoft / microsoft/semantic-kernel

Java: Add AGENTS.md — guidance for AI coding assistants contributing to the repo

Open
#14,028 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

.NET java python triage
Dominant language
C#
Stars
28.6k
Forks
4.8k
Avg merge
14h 13m
Merged PRs (30d)
18

Description

Suggestion: Add an AGENTS.md file

Claude Code, GitHub Copilot, Cursor, and Gemini CLI now read an AGENTS.md file at the repo root to understand how to contribute. It's purpose-built for AI agents, similar to how CONTRIBUTING.md is for humans.

Semantic Kernel is a multi-language AI orchestration SDK — an AGENTS.md could help AI assistants navigate the codebase by covering:

  • Which language SDK to look in for a given concept (Python vs C# vs Java)
  • How to run the test suites in each language
  • Conventions for adding new connectors or memory implementations
  • Which files/patterns to study before modifying the planner or kernel
  • Things to avoid (e.g., breaking cross-language API parity)
Why now

With AI-assisted development becoming the norm, many contributors to projects like this one use Claude Code, Copilot, or Cursor. A well-written AGENTS.md reduces incorrect assumptions and improves the quality of AI-generated PRs.

Projects like LangChain, Pydantic, and the OpenAI SDK have already adopted this convention.

Happy to contribute a draft if that would help.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading CONTRIBUTING.md and reviewing the repository structure for the Python, C#, and Java SDKs. Draft the root AGENTS.md around language selection, test commands, connector and memory conventions, planner or kernel files to study, and cross-language API parity; done means the guidance is complete and useful for AI-assisted contributors.

Written by the indexing model from the issue text.

Assessment

Tech stack
csharp, java, python
Domain
developer-experience, documentation
Issue type
Documentation
Difficulty
3/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.