skills/afelipeg/anthropic-skills-for-enterprise-marketing-os/agency-request-intake-router/Gen Agent Trust Hub
agency-request-intake-router
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a local Python script (
scripts/intake_classifier.py) to classify incoming text requests. The script performs simple keyword matching and returns a structured JSON object. It does not invoke any system shells, perform file operations, or use dangerous execution functions. - [PROMPT_INJECTION]: The skill processes user-provided client requests. While this creates a surface for indirect prompt injection, the risk is negligible as the skill only provides routing recommendations and visual artifacts. It does not possess capabilities like network access or file system modification that could be abused via injection.
- [EXTERNAL_DOWNLOADS]: No external URLs, remote scripts, or third-party dependencies are fetched or executed by the skill.
Audit Metadata