The Agent Skills Directory

[SAFE]: The skill provides instructional content and code templates for software testing. All analyzed files are consistent with the stated purpose of assisting in QA workflows.- [CREDENTIALS_UNSAFE]: The reference documentation (e.g., references/e2e-testing.md, references/performance-testing.md) includes example credentials for test users. These are clearly marked as placeholders for testing environments and do not pose a security risk to the agent or host.- [EXTERNAL_DOWNLOADS]: The skill references and provides installation commands for well-known, trusted testing tools and libraries such as Playwright, k6, and Jest. It also includes snippets for official GitHub Actions, which are standard in modern development pipelines.- [COMMAND_EXECUTION]: The instructions contain numerous examples of CLI commands (e.g., npm, npx, pytest, k6) used to run tests and manage development environments. These are contextually appropriate for a testing specialist skill.- [PROMPT_INJECTION]: The skill defines a workflow to analyze external code and test results, which represents a surface for indirect prompt injection. Ingestion points: Workflow step 4 in SKILL.md (collecting results) and various references to coverage analysis. Boundary markers: Structured output templates in SKILL.md provide clear formatting for reports. Capability inventory: Tasks are restricted to analysis and report generation; no dangerous execution or network logic is triggered by processing these inputs. Sanitization: Not explicitly defined in the provided instructions.

test-master