The Agent Skills Directory

[PROMPT_INJECTION]: The skill facilitates browser automation to interact with and read content from web applications. This creates an indirect prompt injection surface where instructions embedded in the tested web pages or logs could influence the agent's behavior. \n
Ingestion points: browser content accessed via Page.GotoAsync and logs captured via Page.Console in SKILL.md and references/workflows.md. \n
Boundary markers: The provided C# code snippets do not include explicit delimiters or instructions to ignore embedded commands. \n
Capability inventory: The skill allows the use of Bash, Write, and Edit tools, which could be exploited if an injection occurs. \n
Sanitization: No sanitization of scraped browser content is demonstrated in the patterns. \n- [EXTERNAL_DOWNLOADS]: The CI/CD and setup documentation includes instructions for downloading browser binaries and necessary system dependencies. \n
Evidence: The GitHub Actions configuration in references/workflows.md executes playwright.ps1 install to fetch Chromium and other dependencies. This is documented as a standard setup step from a well-known vendor. \n- [COMMAND_EXECUTION]: The skill requires running various CLI tools to build and execute the test suite. \n
Evidence: instructions in SKILL.md and references/workflows.md include the use of dotnet build, dotnet test, and dotnet run, as well as a local script ./run-e2e-tests.ps1 to manage the testing lifecycle.

playwright