The Agent Skills Directory

[SAFE]: No security issues were detected. The skill instructions are focused on legal triage logic and business risk assessment.
[COMMAND_EXECUTION]: The skill includes a local Python script scripts/check-evals.py used for maintenance and self-testing. This script is safe as it only uses the Python standard library and performs static validation of the skill's internal evaluation files without executing external code or performing network requests.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill does not access sensitive system paths or hardcode any credentials. It defines a neutral protocol for fetching legal information from external tools, which is standard for its described functionality.
[PROMPT_INJECTION]: There are no attempts to override agent instructions or bypass safety filters. The instructions are professional and focused on the intended legal triage use case.
[INDIRECT_PROMPT_INJECTION]: The skill processes user legal queries but employs structured output templates and a retrieval protocol that effectively separates user input from the agent's logic, minimizing injection risks. (Severity: SAFE)

opc-legal-counsel