The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run across several scripts (scripts/_skill_protocol_runtime.py, scripts/diagnosis_workspace.py, and scripts/diagnosis_bundle.py) to execute git and its own internal Python scripts. These operations are used for repository analysis and workspace setup and are strictly confined to the target repository and declared workspace directories.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the target repository, such as source code and history. To mitigate injection risks, it includes specific defensive instructions in references/delegation-protocol.md that command the agent to treat repository content as untrusted and to ignore any instructions found within it.
[DYNAMIC_EXECUTION]: The skill utilizes importlib.import_module in scripts/_skill_protocol_runtime.py and implements a directory traversal search for its runtime in scripts/cli.py. These are legitimate uses for modular code loading within the skill's own package structure.

diagnose