competitor-intelligence

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from external social media accounts without appropriate safeguards.
  • Ingestion points: Data enters the agent context through tools such as x_get_profile (bios), x_get_tweets (tweet content), and x_get_followers (follower bios) as described in the workflow section of SKILL.md.
  • Boundary markers: The skill does not define any delimiters or instructions to the agent to disregard commands embedded within the retrieved social media content.
  • Capability inventory: The skill utilizes multiple MCP tools and browser scripts (located in the src/ directory) to process and compare account data.
  • Sanitization: There is no evidence of content validation, escaping, or filtering to prevent malicious instructions in bios or tweets from influencing agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 09:53 AM