The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes existing security check logic and metadata from the local filesystem to maintain consistency when creating new checks. This presents a theoretical surface where malicious instructions in existing source code could influence the agent's generation process.
Ingestion points: Local file reads from prowler/providers/{provider}/services/{service}/.
Boundary markers: None identified; the agent reads files as standard project context.
Capability inventory: The skill has Write access to create files and Bash access to execute the Prowler CLI.
Sanitization: No content sanitization is described for the ingested source code.
[COMMAND_EXECUTION]: The skill uses the Bash tool to verify and run security checks locally through the Prowler CLI.
Evidence: uv run python prowler-cli.py {provider} --list-checks and uv run python prowler-cli.py {provider} --check {check_name} are used for development workflow verification.

prowler-sdk-check