harvest-monitor

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of processing untrusted web content.
  • Ingestion points: The skill uses WebFetch to ingest arbitrary content from external URLs provided by the user or stored in a watchlist.
  • Boundary markers: The instructions do not define clear delimiters or include safety warnings to the agent to disregard instructions that may be embedded within the fetched HTML or text content.
  • Capability inventory: The skill has access to Bash, Write, and Read tools. If an attacker-controlled website contains malicious markdown or commands, the agent might attempt to execute them or exfiltrate local data while performing the 'Impact Assessment' phase of its report.
  • Sanitization: There is no mention of sanitizing or filtering the fetched content before it is processed by the AI for change analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 07:46 AM
Security Audit — agent-trust-hub — harvest-monitor