sales-outbound
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external interest signals to automate CRM entries and email drafting. Ingestion points: Workflow Step 1 in
SKILL.mdprocesses pasted text, Google Sheet data, and Gmail content. Boundary markers: The skill does not define specific delimiters to separate untrusted data from the agent's instructions. Capability inventory: The skill possesses the ability to execute SQL queries (mcp__supabase__execute_sql), draft emails (mcp__google-workspace__draft_gmail_message), and fetch web content (WebFetch). Sanitization:SKILL.mdincludes explicit instructions for SQL literal escaping (Workflow Step 4), and the use of Gmail drafts ensures human review. - [DYNAMIC_EXECUTION]: The skill performs runtime generation of SQL queries by interpolating parsed data into the
references/lead-insert.sqltemplate. The inclusion of escaping logic for single quotes mitigates common injection risks. - [EXTERNAL_DOWNLOADS]: The skill installs tools from the author's GitHub repository (
github.com/peter-tu-zynkr/zynkr-skill-builder) usingnpx.
Audit Metadata