The Agent Skills Directory

[SAFE]: Analysis of the instructions and tool usage patterns reveals no security risks. The skill behaves as a technical guide for an image quality assessment service.
[EXTERNAL_DOWNLOADS]: The documentation includes instructions for users to install additional components from the author's repository (scenario-labs/skills) using the npx utility. This is an expected installation path for the vendor's suite of tools.
[PROMPT_INJECTION]: The skill facilitates a feedback-driven generation loop, which represents a surface for indirect prompt injection. 1. Ingestion points: The agent reads reasons and suggestions from the quality gate tool and project briefs. 2. Boundary markers: Not specified. 3. Capability inventory: Includes scenario_tool_execute_write and model_run. 4. Sanitization: Not explicitly documented. This feedback loop is the primary functional purpose of the skill and is categorized as safe.

scenario-quality-gate