waf-bypass-agent
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, creating an attack surface for indirect prompt injection.
- Ingestion points: The skill explicitly requests untrusted inputs including
blocked_payload,request_context, andresponse_samplesin the 'Inputs' section of SKILL.md. - Boundary markers: The instructions lack explicit boundary markers or delimiters (e.g., XML tags or backticks) to separate untrusted data from instructions, increasing the risk that the LLM might interpret malicious tokens in the
blocked_payloadas instructions. - Capability inventory: The skill describes a workflow for request generation, variant testing, and validation against a
target_endpoint, implying the agent will use network-capable tools to interact with external systems. - Sanitization: There are no instructions provided for sanitizing, escaping, or validating the input data before it is used to generate new payload variants or analyzed for fingerprints.
Audit Metadata