find-skills
Warn
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill facilitates the installation and execution of code from external repositories (such as GitHub) using the
npx skills addcommand. - [EXTERNAL_DOWNLOADS]: Specifically encourages the use of the
-yflag, which skips confirmation prompts and bypasses user review of the package being installed. - [COMMAND_EXECUTION]: Instructs the agent to run shell commands to search for, add, and update packages, which modifies the local environment.
- [PROMPT_INJECTION]: Presents an indirect prompt injection surface by processing untrusted metadata (descriptions and names) of third-party packages retrieved from the network.
- Ingestion points: Output from the
npx skills find [query]command. - Boundary markers: No delimiters or safety instructions are provided to the agent to treat search results as untrusted content.
- Capability inventory: The agent possesses shell execution and file system access capabilities via the CLI commands discussed.
- Sanitization: There is no evidence of filtering or validation of the external content returned by the search command.
Audit Metadata