marketing-seo
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches and installs the
agent-browserdependency from Vercel Labs' official GitHub repository (github.com/vercel-labs/agent-browser). This is a trusted source used for its browser automation capabilities. - [COMMAND_EXECUTION]: The skill executes
agent-browserCLI commands to perform live SERP analysis, technical audits, and schema validation. These commands are parameterised using user-provided keywords and domains. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes text from external websites. 1. Ingestion points: The
agent-browser get text bodycommand inSKILL.mdreads content from third-party sites like Google and SimilarWeb into the agent's context. 2. Boundary markers: The skill does not use specific delimiters or protective instructions to distinguish ingested data from system instructions. 3. Capability inventory: The agent can control a browser and write various audit files to the local directory structure. 4. Sanitization: No sanitization or filtering of the retrieved web content is performed before the LLM processes it.
Audit Metadata