The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process content from various Git submodules and explicitly instructs the agent to read execution commands from a file named AGENTS.md.
Ingestion points: Content within AGENTS.md and the file system under the orgs/** directory structure.
Boundary markers: The skill does not provide delimiters or specific instructions for the agent to ignore potentially malicious embedded instructions in the processed files.
Capability inventory: The skill authorizes the use of shell commands, including submodule, pnpm, bun, and a local script bin/submodules-status.
Sanitization: There are no instructions for validating or sanitizing the content or commands retrieved from the target files.
[COMMAND_EXECUTION]: The skill facilitates the execution of local repository tools and common development environment managers.
It defines workflows for submodule operations (sync, update, status, and a custom smart commit tool with Pantheon integration).
It also utilizes pnpm and bun for package management and development tasks within submodule directories.

submodule-ops