Legal Settlement Evaluation Framework
Security checks across malware telemetry and agentic risk
Overview
This is a document-only, descriptive skill that provides checklists and templates for settlement evaluation and does not request credentials, install code, or instruct the agent to access external systems — its declared requirements align with its stated purpose.
This package is internally consistent and appears to be a safe, document-only framework for evaluating settlements. Before installing or invoking it, consider: (1) Do not paste privileged or confidential client documents into the skill — treat outputs as informational and verify with counsel; (2) remember the skill is not legal advice and you should confirm jurisdiction-specific rules and deadlines with an attorney; (3) because the skill can be invoked by the agent (disable-model-invocation=false), consider whether you want to allow autonomous use in agents that access confidential matter data; disable autonomous invocation if you prefer manual control; (4) monitor future versions or edits to ensure no code or external endpoints are added (the acceptance criteria forbid handlers/scripts — confirm updates respect that). Overall, the package looks coherent with its stated purpose.
SkillSpector
SkillSpector findings are pending for this release.
VirusTotal
No VirusTotal findings