laohan-douyinsousuo
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
openclicommands to perform search operations, verify environment health (doctor), and check login status (whoami). It also runs a local Node.js script located at~/Documents/laohan-skills/laohan-douyinsousuo/scripts/search-evidence.mjsto automate the evidence collection process. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from Douyin search results, which could potentially contain malicious instructions intended to influence the agent's behavior during subsequent processing steps.
- Ingestion points: Untrusted data enters the context via video descriptions and author metadata fetched through
opencliwithin thesearch-evidence.mjsscript. - Boundary markers: While data is structured into JSON and Markdown tables, the skill lacks explicit delimiters or instructions to ignore embedded commands within the ingested text.
- Capability inventory: The skill has access to command execution (via
opencli) and local file system writes (writeFileSyncandrenameSync). - Sanitization: The helper script escapes Markdown pipe characters to prevent table formatting issues, but it does not perform deep sanitization or filtering of the textual content for instruction injection patterns.
Audit Metadata