screaming-frog-tool
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external crawl data (CSV/Excel) and schema blocks, which presents a surface for indirect prompt injection if the source website contains malicious instructions.\n
- Ingestion points: The agent processes user-uploaded CSV/Excel files containing website crawl data (titles, metas, H1s, schema) as described in
SKILL.md.\n - Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the skill's logic.\n
- Capability inventory: The skill documentation describes data interpretation and does not explicitly list subprocess execution, file writing, or network operations, though it mentions the agent can 'trigger crawls' if an MCP server is connected.\n
- Sanitization: No sanitization or validation logic for the ingested crawl data is defined within the skill instructions.
Audit Metadata