lead-enrichment
Warn
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The script disables TLS/SSL certificate validation globally and for its database connection. Evidence in scripts/lead-enrichment.js: process.env.NODE_TLS_REJECT_UNAUTHORIZED = '0' and ssl: { rejectUnauthorized: false }. This allows potential Man-in-the-Middle (MitM) attacks that could intercept the INSFORGE_CONNECTION_STRING containing database credentials.
- [DATA_EXFILTRATION]: The skill sends website URLs from the database to the Apify platform for scraping. Evidence in scripts/lead-enrichment.js shows interaction with the apify-client. This transfers local lead data to a third-party service for processing.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection from data extracted from external websites. 1. Ingestion points: Scraped website content in scripts/lead-enrichment.js. 2. Boundary markers: Absent; there are no instructions to disregard malicious content within scraped data. 3. Capability inventory: Database update permissions on the business_leads table. 4. Sanitization: Lacks semantic validation for scraped content strings, though SQL parameters are used for database safety.
- [SAFE]: The database interaction is implemented using parameterized queries, which successfully prevents SQL injection from the untrusted website data.
Audit Metadata