saas-landing
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads a local file (DESIGN.md) and explicitly instructs the agent to allow a section within that file ('Agent Prompt Guide') to override the skill's core instructions. An attacker providing a malicious design file could potentially manipulate the agent's output or behavior.
- Ingestion points: The agent reads the DESIGN.md file from the current working directory as part of its workflow.
- Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present when processing the external file content.
- Capability inventory: The skill is granted the 'file_write' capability to create the index.html landing page.
- Sanitization: The skill does not validate or sanitize the content of the external file before interpreting its instructions.
Audit Metadata