The Agent Skills Directory

[SAFE]: The skill serves as a specialized persona for social work tasks. Analysis of the provided files, including the core instructions in SKILL.md and the detailed references in the references/ directory, shows no evidence of malicious intent or security risk.
[PROMPT_INJECTION]: No prompt injection or jailbreak patterns were identified. The instructions focus on establishing a professional identity and providing structured clinical guidance.
[DATA_EXFILTRATION]: There are no network operations, hardcoded credentials, or instructions to access sensitive local files. The skill operates within the context of the provided text data.
[REMOTE_CODE_EXECUTION]: The skill does not include any scripts, executable code, or dependencies that would allow for remote code execution.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not use any shell execution syntax (!command) or dynamic context placeholders that could lead to command injection.
[INDIRECT_PROMPT_INJECTION]: While the skill is designed to process user-provided case scenarios, it lacks the technical capabilities (such as file writing or network access) that would make it vulnerable to exploitation via indirect injection. It acts purely as a text-based advisory framework.

social-worker