The Agent Skills Directory

[SAFE]: The skill defines a specialized workflow for utilizing the Momentic MCP tools (e.g., momentic_session_start, momentic_preview_step, momentic_test_splice_steps). All instructions are centered around browser automation and test lifecycle management.
[SAFE]: Data handling is limited to test definitions and session-specific environment variables. The use of placeholders like env.USER_EMAIL or env.USER_PASSWORD follows standard testing practices for managing secrets without hardcoding them.
[SAFE]: The skill includes explicit warnings and 'Non-negotiables' that promote safe usage, such as requiring user confirmation before long-running operations and advising against directly editing YAML files.
[SAFE]: Dynamic execution capabilities, such as the JAVASCRIPT step type, are documented as a fallback mechanism for tasks that native steps cannot perform, which is common in testing frameworks and does not pose an inherent risk in this context.

momentic-test