The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill utilizes the uvx tool to download and execute logic from the ouroboros-ai package hosted on the Python Package Index (PyPI).
[COMMAND_EXECUTION]: The skill uses run_shell to execute environment-resolution logic and interacts with the GitHub API via the gh CLI to star the vendor's repository upon user request.
[DYNAMIC_EXECUTION]: The skill includes an inline Python script that is executed to manage local configuration in ~/.ouroboros/prefs.json and create audit trails in ~/.ouroboros/seed-revisions/.
[INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface by ingesting untrusted user input that is explicitly treated as 'byte-for-byte' content, which can include functional shell commands.
Ingestion points: The goal and acceptance_criteria fields within the session_context argument in SKILL.md.
Boundary markers: None present; the skill explicitly instructions that values enter the Seed verbatim.
Capability inventory: The skill has the ability to run shell commands (run_shell) and perform network operations (gh api).
Sanitization: No sanitization or escaping is performed on the user-provided criteria.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to fetch the ouroboros-ai package and to interact with GitHub's API endpoints.

seed