The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to create and execute local shell scripts ({slug}-feedback-loop.sh) and testing frameworks (such as Jest and Playwright) as part of the bug reproduction and verification process.
Evidence: The workflow in SKILL.md and reference.md explicitly details the creation of feedback loops using various CLI tools and custom scripts.
[EXTERNAL_DOWNLOADS]: The agent is instructed to use tools like curl and httpie to interact with application APIs during the diagnostic phase.
Evidence: reference.md provides examples of using curl in bash scripts to verify API status codes and responses.
[SAFE]: All identified behaviors are integral to the primary purpose of bug diagnosis and are governed by a transparent, documented process. The skill emphasizes user confirmation before applying fixes and requires cleaning up diagnostic code (logs) after completion.

cs-diagnose