The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions specify the use of several shell-based development tools. These include git for managing task status, commits, and notes; pytest for test execution and coverage reporting; and ruff and mypy for code quality checks. These are standard operations for code implementation and verification tasks.
[INDIRECT_PROMPT_INJECTION]: The skill workflow involves ingesting information from project-specific files, which introduces a surface area for instructions embedded in data. Ingestion points: The agent reads task lists and project configuration from plan.md, spec.md, and tech-stack.md as described in the implementation playbook. Boundary markers: No explicit delimiters or instructions are provided to help the agent distinguish between data and potential instructions within these files. Capability inventory: The skill leverages the agent's ability to execute shell commands and modify files within the repository. Sanitization: The instructions do not include steps for sanitizing or validating the content read from these external files.

workflow-patterns