instantly-data-handling
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted lead payloads, email content, and webhook data, creating a surface for potential indirect prompt injection attacks where malicious instructions hidden in external data could influence the agent's behavior.
- Ingestion points: Lead payloads, webhook payloads, and email/reply content as referenced in the Current Contract section of SKILL.md.
- Boundary markers: Absent. The instructions do not mandate the use of specific delimiters or isolation markers when interpolating untrusted data into prompts.
- Capability inventory: The skill uses Read, Glob, Grep, WebFetch, Write, and Edit tools to interact with code and files.
- Sanitization: Instruction 4 directs the agent to redact personal identifiers, email bodies, and API keys from logs and evidence.
- [SAFE]: The skill interacts with official Instantly endpoints and refers to the official TypeScript SDK (@instantlyai/sdk) and CLI (@instantlyai/cli). It promotes secure authentication practices by requiring the use of server-side secret managers and prohibiting the persistence of API keys in logs or the file system.
Audit Metadata