The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the gh CLI through Bash to execute code searches and API requests. These commands incorporate variables provided by the user, such as organization names and action paths. The skill instructions guide the agent to perform these operations in a read-only capacity.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external, untrusted sources (GitHub workflow files). Ingestion points: Workflow files retrieved via gh search code. Boundary markers: The skill does not define specific delimiters to isolate repository content from the agent's instructions. Capability inventory: The agent has access to the gh CLI toolset. Sanitization: There are no explicit sanitization steps to filter out malicious instructions hidden within audited workflow files, which could potentially be used to mislead the auditor.

action-audit