The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The scripts/render_page.py utility automatically installs the 'playwright' package and the Chromium browser from official registries if they are not detected in the environment. These are downloads from a well-known and trusted technology provider.
[REMOTE_CODE_EXECUTION]: The skill uses subprocess.run to execute package installation commands (pip install, playwright install) at runtime to ensure the visual verification environment is prepared.
[COMMAND_EXECUTION]: The instructions guide the agent to generate and execute a start.sh script to manage the deployment of the created application, which involves installing system dependencies and starting local web servers.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it processes untrusted requirements from an external file to drive complex operations like system design and code generation.
Ingestion points: The agent is instructed to read requirements and specifications from /workspace/prd.md.
Boundary markers: The instructions lack specific delimiters or guardrails to prevent the agent from following malicious instructions potentially embedded within the PRD document.
Capability inventory: The skill possesses the capability to execute shell commands, write application code to the file system, and perform network requests for testing.
Sanitization: Content from the requirement documents is processed without validation or sanitization before being used to influence the implementation logic.

glmv-prd-to-app