procore-migration-deep-dive
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill represents a potential surface for indirect prompt injection by ingesting data from the Procore API, which could contain untrusted content.
- Ingestion points: The skill makes GET requests to
api.procore.com(documented inSKILL.md). - Capability inventory: The skill is configured with access to
Write,Edit, andBashtools, allowing for file modification and command execution. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the ingested API data.
- Sanitization: No explicit validation or escaping of the API response data is performed before the agent processes it.
- [DATA_EXFILTRATION]: The skill communicates with official Procore endpoints (
login.procore.com,api.procore.com) for legitimate migration tasks. It follows security best practices by utilizing environment variables (PROCORE_CLIENT_ID,PROCORE_CLIENT_SECRET) for credential management rather than hardcoding secrets.
Audit Metadata