clay-sdk-patterns
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides integration patterns for Clay, a well-known SaaS platform. All network communications are directed to the official API domain (api.clay.com).
- [SAFE]: Secret management follows industry best practices by recommending the use of environment variables (CLAY_API_KEY, CLAY_WEBHOOK_URL) for credential handling in the singleton pattern.
- [SAFE]: The documentation explicitly instructs users to redact sensitive information, such as webhook URLs and credentials, from logs to prevent data exposure.
- [INDIRECT_PROMPT_INJECTION]: The skill implements a webhook handler that ingests external data (req.body) and API responses. This represents an attack surface for indirect prompt injection if the resulting data is subsequently interpolated into LLM prompts without proper boundary markers or sanitization.
- Ingestion points: TypeScript client methods (sendToTable, sendBatch) and Python client methods (send_to_table, enrich_person) that handle table data and API responses; Express callback handler in Step 4.
- Boundary markers: No explicit boundary markers or "ignore instructions" warnings are included in the provided code snippets.
- Capability inventory: The skill instructions enable network requests (fetch, httpx) and provide access to shell tools (Bash) and file modification (Write, Edit).
- Sanitization: While the skill suggests log redaction, it does not include logic for sanitizing or escaping the content of records received from Clay before further processing.
Audit Metadata