The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources, specifically Jira issues (Acceptance Criteria and team comments), which creates a potential surface for indirect prompt injection. To mitigate this risk, the instructions explicitly treat these synced files as read-only caches and utilize structured ADF (Atlassian Document Format) conversion for all outbound writes to the issue tracker.
[DYNAMIC_EXECUTION]: The workflow utilizes eval() within the Playwright testing framework to extract DOM element coordinates and dimensions. Additionally, it uses osascript (JavaScript for Automation) to facilitate rich-text reporting for Slack. Both instances are legitimate uses of dynamic execution necessary for UI testing and developer workflow automation.
[COMMAND_EXECUTION]: The skill relies on a suite of standard command-line tools such as bun, git, gh, curl, acli, and xray-cli to perform project management, code exploration, and API testing. These operations are performed within the expected scope of a QA and sprint orchestration utility.
[EXTERNAL_DOWNLOADS]: The skill references configurations and code repositories from trusted organizations, including Anthropic and Vercel Labs. These references are documented neutrally as part of the skill's operational dependencies.

sprint-testing