The Agent Skills Directory

[COMMAND_EXECUTION]: The skill extensively uses the GitHub CLI (gh) to perform API calls, check authentication status, and trigger remote workflows using gh workflow run. Additionally, it executes a local Node.js script (scripts/fleet-scorecard.mjs) to aggregate fleet-wide metrics.
[INDIRECT_PROMPT_INJECTION]: The skill ingests content from managed child repositories, including cron-state.json, aeon.yml, and metadata from remote SKILL.md files. This creates a surface where manipulated data in a child repository could potentially influence the agent's reporting or logic during execution.
Ingestion points: Fetches file contents (memory/cron-state.json, aeon.yml) and repository metadata from remote managed instances via the GitHub API.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing the content retrieved from remote repositories.
Capability inventory: The skill possesses significant capabilities, including triggering remote GitHub Actions workflows, executing local scripts, and sending notifications.
Sanitization: There is no mention of specific sanitization, escaping, or validation logic for the content fetched from remote repositories before it is processed or reported.
[EXTERNAL_DOWNLOADS]: The skill fetches data and file contents from GitHub's official API (api.github.com). As GitHub is a well-known service, these operations are documented as part of the skill's core functionality for managing a fleet of repositories.

fleet-control