The Agent Skills Directory

[PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by reading work package specification files (wps.md) and using their content to generate implementation instructions for the agent.
Ingestion points: The skill reads specs/<SPEC_FOLDER>/wps.md to identify work packages and context.
Boundary markers: Absent; the execution prompt in exec-prompt.md does not utilize delimiters or warnings to isolate ingested file content from the agent's instructions.
Capability inventory: The skill uses Glob and Read tools, and the final prompt triggers code implementation tasks including file modifications.
Sanitization: No sanitization or validation of the ingested file content is performed.
[COMMAND_EXECUTION]: The skill suggests several shell commands for Git repository management, including git pull, git submodule update, and git checkout. These commands are provided as text for the user to run manually to ensure the environment is prepared.

lf-exec