The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use standard shell utilities such as cat, grep, and md5 to inspect team state and verify synchronization. It also involves Git operations (git pull, git push, git fetch) which are necessary for the skill's primary purpose of multi-person coordination.
[EXTERNAL_DOWNLOADS]: The workflow relies on synchronizing with a remote repository via git pull and git fetch. These operations target the official project repository ('origin main').
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it reads and reports on data from untrusted or shared sources within the repository. Specifically:
Ingestion points: The agent is instructed to read _project_specs/team/state.md, _project_specs/todos/active.md, and _project_specs/team/contributors.md at the start of a session.
Boundary markers: There are no defined delimiters or instructions to ignore embedded commands within these shared data files.
Capability inventory: The agent has the ability to execute shell commands (git, cat, grep), write to the filesystem, and push changes to a remote repository.
Sanitization: The skill does not describe any validation or sanitization of the content found in the shared markdown files before processing or reporting it to the user.

team-coordination