resend-api
Warn
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [METADATA_POISONING]: The frontmatter in SKILL.md claims the author is OpenAI, which is deceptive as the skill is authored by an external party. This could lead users to misjudge the trustworthiness and safety of the skill.
- [INDIRECT_PROMPT_INJECTION]: The skill provides tools and instructions for processing inbound emails via the Resend Receiving Emails API, creating a surface for indirect prompt injection attacks where malicious instructions in received emails could affect agent logic.
- Ingestion points: Inbound email metadata and content retrieved from the Receiving Emails API in scripts/resend_api.py.
- Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore potentially malicious content within emails.
- Capability inventory: The skill allows the agent to manage domains, API keys, and send emails via scripts/resend_api.py.
- Sanitization: No instructions are provided for sanitizing inbound email content before it is processed by the agent.
- [DATA_EXFILTRATION]: The bundled helper script scripts/resend_api.py makes network requests to api.resend.com. This is intended functionality for communicating with the official API of Resend, a well-known technology service.
Audit Metadata