The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the provided skill files.
[DATA_EXFILTRATION]: The skill instructs the agent to use network search for verifying facts (e.g., in SKILL.md and sop-person.md). This is a standard functional requirement for research and analysis and does not involve accessing sensitive local files or credentials.
[REMOTE_CODE_EXECUTION]: The analysis confirms there are no instances of remote script downloads, package installations, or dynamic code execution like eval() or exec().
[PROMPT_INJECTION]: The skill's instructions are focused on analysis frameworks and do not contain attempts to bypass safety filters, extract system prompts, or override agent constraints.
[INDIRECT_PROMPT_INJECTION]: While the skill processes external data via web search (Category 8 surface), the SOPs explicitly guide the agent to cross-reference multiple sources and prioritize primary data, which serves as a logical mitigation for processing potentially untrusted web content.

ls-analyze