qiaomu-smart-search
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is designed to interact with the host system via the opencli command-line tool. It instructs the agent to run multiple subcommands to discover available platforms, read live help documentation, and perform search queries.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data retrieved from external platforms (such as Google, Twitter, and various news sites) via opencli. This content is introduced into the agent's context, where it could potentially contain malicious instructions intended to manipulate the agent's behavior.
- Ingestion points: Search results from numerous external platforms fetched through the opencli utility (referenced in SKILL.md and throughout the reference files).
- Boundary markers: The skill does not provide instructions to use delimiters or specific safety prompts to isolate untrusted web content from its own instructions.
- Capability inventory: Shell command execution via the opencli binary, which is used repeatedly for discovery and searching.
- Sanitization: There are no requirements or steps provided to sanitize, filter, or validate the results obtained from external sources before the agent processes them.
Audit Metadata