The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to load and follow instructions from an external file located at @$HOME/.claude/get-shit-done/workflows/ship.md. This creates a dependency on untrusted data if that file can be influenced by other scripts or users on the system.
Ingestion points: The contents of $HOME/.claude/get-shit-done/workflows/ship.md are ingested into the agent's context as the primary execution instructions.
Boundary markers: The skill does not define specific delimiters or instructions to prevent the agent from following malicious commands that might be embedded in the workflow file.
Capability inventory: The skill possesses powerful capabilities including the Bash, Write, and Read tools, which could be misused if the loaded workflow is compromised.
Sanitization: There is no mechanism described to validate or sanitize the workflow file before its instructions are executed.
[DYNAMIC_EXECUTION]: The skill dynamically resolves a file path using the $HOME environment variable to load its execution logic at runtime rather than having its instructions self-contained.

gsd-ship