subkeyword-injector
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external websites via the Browser MCP and Search Console queries, which could be used for indirect prompt injection attacks.
- Ingestion points: Web content extraction via Browser MCP (references/page-access.md) and Google Search Console query data (SKILL.md).
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are defined for the ingested content.
- Capability inventory: The agent has the capability to modify local files (SKILL.md, step 4) if authorized by the user.
- Sanitization: No explicit sanitization or validation of the fetched content is described before it is used to generate or apply edits.
Audit Metadata