The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses draft-cli commands to manage the review lifecycle.
Commands like draft status, draft open, and draft comments list are utilized to interface with the Draft platform.
These operations are consistent with the skill's stated purpose of facilitating a local-to-remote review loop.
[EXTERNAL_DOWNLOADS]: The skill documentation specifies a requirement for the @innosage/draft-cli Node.js package.
This package is a vendor-owned resource associated with the author, innosage-llc.
[DATA_EXFILTRATION]: Local workspace markdown content is transmitted to the Draft service when the draft open command is executed.
This transmission is the primary intended function of the skill to enable external human review on the Draft GUI.
[PROMPT_INJECTION]: The skill processes external data by reading comments from the Draft service.
Ingestion points: Comments are retrieved via draft comments list <path> --json (SKILL.md).
Boundary markers: The skill does not explicitly define delimiters or instructions for the agent to ignore potentially malicious instructions within comments.
Capability inventory: The agent is instructed to use file-editing tools to apply feedback to local files.
Sanitization: No specific sanitization or validation steps are defined for the ingested comment data.

draft-review-loop