dianping-info-query

Pass

Audited by Gen Agent Trust Hub on Mar 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it extracts and processes untrusted user-generated content from a public website.
  • Ingestion points: Business reviews, user tags, and shop descriptions are extracted from dianping.com using the readability and snapshot actions as described in SKILL.md and references/deep-info-extraction.md.
  • Boundary markers: The skill does not define specific boundary markers or instructions to isolate extracted web content, increasing the risk that the agent might follow instructions embedded in a user review.
  • Capability inventory: The skill utilizes powerful browser interaction capabilities, including navigate, click, type, and list_tabs, which could be exploited if an injection is successful.
  • Sanitization: No sanitization or filtering logic is present in the skill to clean external text before the agent interprets it.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 19, 2026, 12:03 PM