The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to create directories, copy files, and perform git operations. Specifically, it executes mkdir -p, cp, and git commit to manage the Shipyard project state.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from various external markdown files (e.g., spec.md, plan.md, tasks.md) and incorporates their content into project definitions and subagent prompts.
Ingestion points: Files located in the user-provided <feature-path>, including spec.md, plan.md, research.md, data-model.md, tasks.md, and the contracts/ directory, as well as .specify/constitution.md.
Boundary markers: The skill does not use specific delimiters or instructions to ignore embedded commands when reading from these files or passing their content to the architect subagent.
Capability inventory: The skill can perform file system modifications (writes and directory creation), execute shell commands via git, and dispatch subagents (shipyard:architect).
Sanitization: There is no evidence of sanitization, validation, or escaping of the content ingested from the external markdown files before it is processed or stored.

import-spec