lit-to-bazel
Skill948 starsChanged 7 months agoApache-2.0View on GitHub
Converts MLIR lit test files to `bazel run` commands. Use when you need to convert a failing MLIR lit test target (containing RUN lines) to a shell command whose flags can be modified for further analysis and debugging.
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- dump-intermediate-irSkill
- e2e-to-bazelSkill
- reproduce-failureSkill
The file
Files
Lit To Bazel
Overview
This skill guides the agent in using the lit_to_bazel tool to convert MLIR
test files to executable commands.
Usage
Converting a Test File
To convert a lit test file to a bazel run command, use the following command
recipes:
bazel run //third_party/heir/scripts:lit_to_bazel -- {test_file_path}Replace {test_file_path} with the absolute path to the test file. The absolute
path is necessary because the bazel run command runs relative to its runfiles
directory, and so a relative path will not be relative to the VCS root of the
project.
Gotchas
- Absolute Paths: The tool replaces
%sin RUN lines with the absolute path of the test file to ensure it works in the sandbox. - Piped RUN commands: The tool supports piped RUN lines, such as piping heir-opt into heir-translate.
- Complex RUN commands: Some very complex RUN lines are not supported, such as lit test files with multiple RUN lines that interact via writing to and reading from temporary files. When encountering situations like this, you should stop and warn the user.
- Non-lit tests: Some test targets are not lit tests, and these do not need
the
lit_to_bazeltool.
Copy this checklist and track progress:
- [ ] Step 1: Identify the test file to convert.
- [ ] Step 2: Run the `lit_to_bazel` tool on the file.
- [ ] Step 3: Inspect the generated command or run it.
- [ ] Step 4: Verify execution results.<!-- mdformat global-off -->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
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.

