compose-outreach
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill is designed to retrieve and process data from external sources, including Common Room signals and web search results, to generate outreach drafts. This pattern introduces a potential for indirect prompt injection if the external content contains malicious instructions intended to influence the agent's behavior.
- Ingestion points: External data enters the agent's context through Common Room MCP tools and web searches described in
SKILL.md(Step 1 and Step 2). - Boundary markers: The instructions do not define explicit delimiters or 'ignore' commands to separate untrusted external data from the agent's primary instructions.
- Capability inventory: The skill's functionality is limited to text generation. There are no identified scripts or tools within the skill that allow for arbitrary command execution, file system writes, or persistence mechanisms, which significantly limits the potential impact of an injection.
- Sanitization: There are no documented steps for validating, filtering, or sanitizing the data retrieved from external tools before it is used to compose the outreach messages.
Audit Metadata