discoverability-growth
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill performs automated crawling of external websites (including competitors) using
advertools. This content (titles, headers, and metadata) is ingested into the AI's context for synthesis and backlog generation, creating a surface for indirect prompt injection if a targeted website contains malicious instructions hidden in its HTML. - Ingestion points:
advertoolscrawl output from the current site and competitor sites (Step 3 and Step 5). - Boundary markers: Absent. The skill instructions do not specify the use of delimiters or 'ignore' instructions for the crawled data.
- Capability inventory: Access to
Bash(command execution),Write(file system access), andEdit(file modification). - Sanitization: Absent. Extracted string data from the crawler is passed directly to the LLM for topic clustering and intent analysis.
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to execute Python code via thepython -cpattern to perform SEO tasks. While these commands are defined within the skill, dynamic script generation at runtime increases the potential impact of command injection if user-supplied inputs (like URLs or keywords) are not properly escaped. - [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install external Python packages (
advertoolsandseo-keyword-research-tool) at runtime usingpip. While these are established SEO tools, runtime installation from public registries (PyPI) introduces a standard dependency management risk.
Audit Metadata