phishing-campaign-builder
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow relies on researching 'current public context' and 'org voice' through tools like
pretext-brainstormerandphishing-pretext. This process involves ingesting untrusted data from the internet (such as social media, company websites, or public forums) which creates a surface for indirect prompt injection. - Ingestion points: External research into target organization culture, department names, and public communication styles (referenced in
SKILL.mdsteps 1 and 3). - Boundary markers: The skill lacks explicit delimiters or instructions to treat researched content as untrusted data, increasing the risk that embedded instructions in external sources could manipulate the agent's behavior.
- Capability inventory: The skill is capable of generating structured HTML templates, plain-text email bodies, and landing page copies based on the gathered research.
- Sanitization: No mechanisms are provided to validate or sanitize the data retrieved from external sources before it is interpolated into the phishing email templates.
Audit Metadata