n8n-workflow-design
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The instructions suggest the command
npx -y @coldiq/mcp@latestto connect the ColdIQ MCP. This command downloads and executes a package from the npm registry at runtime. As the package namespace@coldiqmatches the skill's author, this is categorized as legitimate vendor tool installation. - [EXTERNAL_DOWNLOADS]: The skill references the vendor's domains
coldiq.comandapi.coldiq.comfor obtaining API keys and performing service integrations. These are identified as vendor-owned resources and facilitate the skill's primary purpose. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from a local file
{SKILL_BASE}/resources/n8n-core-guide.md. This represents a potential surface for indirect instructions, although no specific malicious patterns were detected in the primary skill file. - Ingestion points:
{SKILL_BASE}/resources/n8n-core-guide.md(read as a reference guide). - Boundary markers: Absent in the analyzed SKILL.md.
- Capability inventory: Suggests shell command execution (npx) and HTTP node configuration for API calls.
- Sanitization: Not explicitly present for the referenced guide.
Audit Metadata