seo-content-blog
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external URLs (competitors, user sites) to perform audits and generate content. This creates an attack surface for indirect prompt injection where malicious instructions on those external pages could attempt to influence the agent's behavior.
- Ingestion points: Competitive benchmarking (
page_benchmark.py) and site audits (seo_audit.py) ingest and analyze data from arbitrary, attacker-controlled URLs. - Boundary markers: No explicit markers or delimiters are defined to separate the untrusted web content from the agent's core instructions.
- Capability inventory: The agent has the capability to execute shell commands (
python3), read and write local files (Obsidian vault), and perform network operations via crawling scripts. - Sanitization: The instructions do not specify any sanitization, filtering, or validation of the content retrieved from external sources before it is processed by the LLM.
- [COMMAND_EXECUTION]: The skill relies on the agent's ability to execute local Python scripts via the shell (e.g.,
seo_audit.py,page_benchmark.py,site_crawl.py). While this is the intended primary purpose of the skill, it constitutes a high-privilege capability that increases the potential impact if an injection attack occurs.
Audit Metadata