seo-content-data-study
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external and untrusted data from multiple sources, which could be leveraged for indirect prompt injection attacks.
- Ingestion points: The skill ingests data from CSV exports via the
scripts/data_study.pyscript and reads local knowledge bases, such as Obsidian vaults and markdown notes, as instructed inSKILL.md. - Boundary markers: The instructions do not define or utilize explicit boundary markers or delimiters to separate ingested data from agent instructions, nor do they include warnings to ignore embedded commands.
- Capability inventory: The skill environment supports the execution of shell commands (specifically
python3) and provides the agent with read and write access to the local file system. - Sanitization: There is no evidence of input sanitization, filtering, or validation to identify or neutralize malicious instructions that might be embedded within the processed data files.
Audit Metadata