agentleFS
Sign inStart free
zilliztechzilliztech/knowhere/AGENTS.md

knowhere

AGENTS.md390 starsChanged 10 days agoApache-2.0View on GitHub

Knowhere is the C++ vector search library used by Milvus. Public interfaces live in include/knowhere/, index implementations in src/index/, and unit tests in tests/ut/. Maintain project instructions in this file and task-specific guidance in .agents/skills/. CLAUDE.md imports this file. Read task-specific guidance through the skill links below, and edit the shared files so agents receive the same guidance. Read the relevant skill before working on its topic: - Run commands from the repository root. Use make help for build options.…

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,542 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository7 other files it gives its agents.

The file

We split this AGENTS.md into the rules it gives an agent

7 rules. About 0 are common, 2 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the relevant skill before working on its topic:

    line 12
  2. Very specific:

    Run commands from the repository root.

    line 25
  3. Very specific:

    Follow .clang-format: Google C++ style, 120-column limit, four-space indentation.

    line 28
  4. Very specific:

    Run pre-commit run --files <changed-files> before committing.

    line 29
  5. Very specific:

    Verify commands and supported features against the current Makefile, conanfile.py, CMake files, index registrations, and tests.

    line 30
  6. Distinctive:

    Keep changes scoped and preserve unrelated local work.

    line 32
  7. Very specific:

    Keep CLAUDE.md as an @AGENTS.md import.

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Knowhere

Knowhere is the C++ vector search library used by Milvus. Public interfaces live in include/knowhere/, index implementations in src/index/, and unit tests in tests/ut/.

Shared context

Maintain project instructions in this file and task-specific guidance in .agents/skills/. CLAUDE.md imports this file. Read task-specific guidance through the skill links below, and edit the shared files so agents receive the same guidance.

Read the relevant skill before working on its topic:

TopicSkill
Source layout, index implementations, dependenciesarchitecture
Build configuration and compiler errorsbuilding
Running and adding teststesting
Index performance optimization and benchmarksindex-optimization
Formatting and static analysiscode-quality
Commits, PR titles, DCO, and labelsgit-workflow

Development

  • Run commands from the repository root. Use make help for build options.
  • Build CPU unit tests with make WITH_UT=True; the default executable is build/Release/tests/ut/knowhere_tests. Select relevant tests by name or Catch2 tag.
  • Follow .clang-format: Google C++ style, 120-column limit, four-space indentation. Run pre-commit run --files <changed-files> before committing.
  • Verify commands and supported features against the current Makefile, conanfile.py, CMake files, index registrations, and tests. Update skill guidance when those sources change.
  • Keep changes scoped and preserve unrelated local work. Report the checks actually run and any validation still pending. Follow the user's requested boundary for committing, pushing, and creating PRs.

Maintaining the shared layout

Keep CLAUDE.md as an @AGENTS.md import. Add new skills under .agents/skills/ and update the table above. Keep relative references inside a skill with that skill.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free