web-pentest

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.72). The skill’s runtime workflow fetches and ingests outsider-authored free text from public web content during Phase 2 recon (e.g., curl of robots.txt, sitemap.xml, and .well-known/* plus browser navigation/XHR capture), which can contain attacker-controlled prompt-injection strings that get placed into the agent’s LLM context via the evidence/analysis pipeline.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)