The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to utilize local development and build tools (e.g., dotnet build, pytest --collect-only, mvn test-compile, go vet) to verify that added test tags do not break the project's build or discovery process. It also provides fallback instructions to use standard shell commands like cat, sed, or Get-Content for reading and writing files if primary agent tools encounter environment-specific path or transport issues.
[INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface where untrusted data is processed with write and execute capabilities.
Ingestion points: Test files and project structures from the local workspace.
Boundary markers: The instructions lack explicit boundary markers or warnings to ignore embedded instructions in the ingested code.
Capability inventory: The agent has the ability to write to source files (tagging) and execute shell commands/build tools (verification).
Sanitization: No explicit sanitization of external content is mentioned before processing.

test-tagging