pricing
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes a feature to perform 'teardowns' of external pricing pages by reading their content from URLs or pasted text. This creates an indirect prompt injection surface where instructions hidden on a third-party website could potentially influence the agent's output.\n
- Ingestion points: The
references/pricing-page-teardown.mdfile defines the process for auditing live pricing pages.\n - Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the external pricing page data.\n
- Capability inventory: The skill's capabilities are restricted to generating textual analysis and reports; it does not demonstrate capabilities for file system modification, credential access, or arbitrary network requests.\n
- Sanitization: There is no evidence of content sanitization or filtering applied to the ingested external data.\n- [NO_CODE]: The skill consists entirely of markdown and JSON configuration files. No executable scripts (such as Python, JavaScript, or shell scripts) are included, which significantly reduces the attack surface for direct command execution, privilege escalation, or persistence attacks within the host environment.
Audit Metadata