blitz-list-builder
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The
scripts/find-contacts.tsscript performs network requests toapi.useblitz.com(or a user-definedBLITZ_BASE_URL) to fetch employee and company data using a provided API key. - [COMMAND_EXECUTION]: The skill instructions direct the user to execute a local TypeScript script using
npx tsx. This is a standard method for running development scripts but involves executing code with file system and network permissions. - [CREDENTIALS_UNSAFE]: The skill requires a
BLITZ_API_KEYenvironment variable. It correctly avoids hardcoding secrets and documentation demonstrates standard environment variable usage. - [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes data from an external API (Blitz) and user-supplied domain lists that are later used as input for AI-based qualification tools like
/icp-prompt-builder. - Ingestion points:
scripts/find-contacts.tsprocesses input from files provided via--domains-fileand data returned from the Blitz enrichment API. - Boundary markers: No specific delimiters or instructions to ignore embedded content are used in the CSV output, which is the format intended for subsequent AI analysis.
- Capability inventory: The script has the capability to perform network requests (
fetch) and write to the local file system (writeFileSync). - Sanitization: Input domains are cleaned using a
cleanDomainhelper, but data retrieved from the external API (such as job titles or company descriptions) is saved to the CSV without specific sanitization against injection patterns.
Audit Metadata