The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process external project documentation and steering files to inform its testing strategy and code generation. This creates an attack surface where malicious instructions embedded in those documents could influence the agent's behavior.
Ingestion points: The agent reads configuration and requirements from steering/ (e.g., structure.md, tech.md, product.md) and docs/requirements/ directories.
Boundary markers: There are no specific delimiters or instructions provided to the agent to ignore potentially malicious embedded prompts within these external files.
Capability inventory: The agent has access to the Bash tool for command execution and the Write/Edit tools for modifying the filesystem.
Sanitization: No explicit sanitization or validation of the ingested content is performed before it is used to generate code or run commands.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform essential testing tasks and manage project workflows.
The agent executes test runners such as npm run test and uv run pytest.
It interacts with specialized CLI tools like musubi-workflow for workflow stage management and musubi-browser for automated browser testing.
These operations are consistent with the skill's primary purpose and involve standard developer tooling and environment interaction.

test-engineer