send-notification
Pass
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing untrusted data. \n
- Ingestion points: Data is ingested from user-provided 'Notification content', external 'Related context' (such as campaign IDs or metric snapshots), and local configuration files located in
~/.claude-marketing/(SKILL.md). \n - Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat external content as data rather than instructions (SKILL.md). \n
- Capability inventory: The skill has the capability to read local files in the
~/.claude-marketing/directory and perform network write operations to Slack and Intercom via MCP servers (SKILL.md). \n - Sanitization: The process lacks documented sanitization, escaping, or validation steps for the external content before it is summarized and sent to messaging platforms.
Audit Metadata