The Agent Skills Directory

External Data Processing: The skill is designed to process external data from GitHub issues and user instructions to guide its reproduction logic. This creates a potential surface for indirect prompt injection where instructions embedded in an issue could attempt to influence the agent. To manage this, the skill explicitly instructs the agent to treat user-provided hints as secondary to the core issue description and strictly follow provided identifiers.
Isolated Command Execution: Operations involving tools like git, npm, and wrangler are performed within a dedicated container backend. This ensures that the reproduction environment is isolated from the agent's primary shell and the workspace filesystem, minimizing the risk of accidental data access or system modification.
Trusted Infrastructure Integration: The skill interacts exclusively with well-known services, such as GitHub and Cloudflare's temporary deployment flow. It utilizes secure deployment methods that do not require hardcoded credentials, and all external downloads originate from standard package registries and official repositories.

reproduce