m26-resend-email
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill requires the agent to clone the 'm26pipeline' repository from GitHub and execute various Python scripts (e.g., 'send_campaign.py') locally. This enables the execution of code fetched from a remote source.
- [COMMAND_EXECUTION]: The skill uses shell commands for environment setup, dependency management, and script execution, specifically including 'git', 'pip', and 'python'.
- [EXTERNAL_DOWNLOADS]: The skill downloads project source code from GitHub and installs necessary Python packages from public registries.
- [PROMPT_INJECTION]: The skill processes untrusted external data, which creates a surface for indirect prompt injection. 1) Ingestion points: Incoming Gmail replies retrieved by 'sync_replies_gmail.py' and campaign data from Google Sheets processed by 'send_campaign.py'. 2) Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions embedded in the external content. 3) Capability inventory: The skill has the capability to send emails via Resend, update Google Sheets via webhooks, and execute shell scripts. 4) Sanitization: No data sanitization or validation mechanisms are described for the external inputs before they are incorporated into the agent's workflow.
Audit Metadata