sendgrid-email-log
Pass
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The bash scripts scripts/get_email_log.sh and scripts/query_email_logs.sh execute embedded Python code via a heredoc pattern for local data processing and JSON manipulation.
- [CREDENTIALS_UNSAFE]: The skill handles SendGrid API keys, allowing users to pass them as arguments or store them in a local configuration file (~/.config/sendgrid-send-email/api_key). It follows security best practices by setting restricted file permissions (chmod 600) on the stored credentials.
- [DATA_EXFILTRATION]: The skill transmits the API key to SendGrid's official domains (api.sendgrid.com and api.eu.sendgrid.com) to authenticate log retrieval requests. These are well-known services and the activity is limited to legitimate functionality.
- [PROMPT_INJECTION]: The skill processes external data retrieved from SendGrid API responses, representing a potential ingestion surface for indirect prompt injection. Ingestion points: scripts/get_email_log.sh and scripts/query_email_logs.sh; Boundary markers: None present; Capability inventory: Network access via curl and local file writing; Sanitization: The data is parsed as JSON, but string content is not sanitized for LLM safety.
Audit Metadata