turbosnap-debug-customer
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to recommend several shell commands to the user for local execution, including
rg(ripgrep) for configuration discovery andgitfor repository state verification. - [EXTERNAL_DOWNLOADS]: Recommends the use of
npx @chromatic-com/turbosnap-helperandnpx chromatic trace, which fetch vendor-owned packages from the official NPM registry to assist in local debugging. - [PROMPT_INJECTION]: The skill involves processing untrusted user data such as pasted logs, CLI output, and configuration files. While this presents an indirect prompt injection surface, the skill is designed to analyze this data for diagnostic purposes rather than for automated code execution or external communication.
- Ingestion points:
SKILL.md(Workflow),reference/intake-templates.md(Log/CLI output intake). - Boundary markers: Absent; the skill relies on natural language parsing of the user's provided snippets.
- Capability inventory:
COMMAND_EXECUTION(suggestinggit,rg, andnpxcommands to the user). - Sanitization: Absent; the skill processes raw text inputs as provided by the customer.
Audit Metadata