deal-strategist
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8). It processes untrusted business data such as account names, deal evidence, and competitive info which is then used to generate assessments. If this data contains malicious instructions, the agent might follow them. Ingestion points: SKILL.md templates for Opportunity Assessment and Competitive Battlecard. Boundary markers: None identified; no specific delimiters or instructions to ignore embedded commands are present. Capability inventory: Access to bash, git, and file system operations. Sanitization: No evidence of input validation or escaping for the processed deal data.- [COMMAND_EXECUTION]: The skill explicitly allows the use of powerful CLI tools including bash and git. These capabilities are intended for project management and logging but increase the impact of a successful prompt injection attack.
Audit Metadata