find-leads
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
curlcommand-line tool to perform GraphQL queries and mutations against the CRM API athttps://crm.psquared.dev/graphql. These operations are used to check for existing leads and register new contacts. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes content from external websites via
WebFetchto determine lead relevance and legal compliance. Ingestion points: Content from external websites fetched during the lead search phase (SKILL.md). Boundary markers: There are no explicit instructions or delimiters used to prevent the agent from following commands that might be embedded in the fetched website text. Capability inventory: The skill can execute network requests viacurland write to the localskip-list.jsonfile. Sanitization: There is no evidence of filtering or sanitization of the external website data before it is interpreted by the agent.
Audit Metadata