The Agent Skills Directory

[SAFE]: The skill was audited for all ten threat categories, including prompt injection, data exfiltration, and obfuscation. No malicious patterns were identified.\n- [NO_CODE]: The skill consists exclusively of Markdown files defining analytical procedures. It does not include scripts, executables, or configurations for external tools, eliminating risks associated with code execution or dependency vulnerabilities.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze user-provided financial evidence. Although this constitutes an ingestion surface for potentially malicious data, the risk is negligible as the skill lacks any capabilities (such as shell access, network operations, or tool invocation) to carry out side effects. The 'Output contract' defined in SKILL.md and the reference files provides structural boundaries that reinforce model focus on the intended analytical task.

investor-lenses