seo-content-brief
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: Recommended installation uses a piped shell script from cli.inference.sh, which is the developer's infrastructure.
- [EXTERNAL_DOWNLOADS]: Fetches CLI binaries and verification files from dist.inference.sh.
- [COMMAND_EXECUTION]: Executes infsh and npx commands to perform research tasks and install skill extensions.
- [PROMPT_INJECTION]: Ingests external web content which presents an indirect injection surface. Evidence: 1. Ingestion: URL extraction in SKILL.md. 2. Boundary markers: None. 3. Capabilities: CLI tool execution in SKILL.md. 4. Sanitization: None.
Audit Metadata