find-skills
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and evaluate instructions and scripts from third-party candidate skills. This creates an attack surface where a malicious candidate skill could embed instructions to subvert the agent during the evaluation or installation process.
- Ingestion points: The agent is directed to read
SKILL.mdand relevant scripts from external candidates found via the registry or theskills.shwebsite. - Boundary markers: There are no instructions defining specific delimiters or boundary markers to separate candidate content from the agent's internal reasoning or system instructions.
- Capability inventory: The skill utilizes shell command execution via
npxand file system operations, including writing files and creating symlinks during the installation phase. - Sanitization: The skill relies on manual agent evaluation of source maintenance, trigger scope, and side effects rather than technical or programmatic sanitization of the external content.
- [EXTERNAL_DOWNLOADS]: The skill uses the
npx skillscommand to interact with an external package registry and referenceshttps://skills.sh/to discover and fetch external skill definitions and scripts. - [COMMAND_EXECUTION]: The skill workflow involves executing shell commands, specifically
npx skills find, to perform searches and manage the installation of external components.
Audit Metadata