skills/reason-machines/marketing-skills/whatsapp-instagram-tiktok-mass-sender-marketing/Gen Agent Trust Hub
whatsapp-instagram-tiktok-mass-sender-marketing
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill extracts untrusted data from social media profiles, such as usernames, bios, and interests, and interpolates this content directly into automated messages and comments.
- Ingestion points: The
scrapeFollowersandgetHashtagUsersfunctions inSKILL.mdingest external user data. - Boundary markers: No boundary markers or instructions to ignore embedded commands are present in the provided templates.
- Capability inventory: The skill utilizes network capabilities via
api.sendDM,api.comment, andapi.whatsapp.sendBulk, alongside file writing viafs.writeFileSyncinSKILL.md. - Sanitization: No evidence of sanitization, escaping, or filtering is present for the user-supplied data before it is processed by the messaging functions.
- [DATA_EXFILTRATION]: The skill is designed to harvest information from social media users, including the extraction of phone numbers from profile bios, which are then transmitted to external APIs for marketing purposes.
- [COMMAND_EXECUTION]: The skill's implementation patterns include local file system operations using the Node.js
fsmodule to persist tracking data for contacted users and opt-outs.
Audit Metadata