The Agent Skills Directory

[SAFE]: No security issues or malicious patterns were identified. The skill performs its described functions using standard system utilities and the official GitHub CLI.
[COMMAND_EXECUTION]: The skill executes local bash scripts that utilize gh, jq, grep, sed, and awk. These commands are used strictly for parsing workflow YAML files and calculating schedule distributions, which is consistent with the skill's intended purpose.
[DATA_EXFILTRATION]: All network operations are directed to the GitHub API via an authenticated gh session. There is no evidence of sensitive data (such as credentials or environment variables) being exfiltrated to external or untrusted servers.
[INDIRECT_PROMPT_INJECTION]: The skill processes external GitHub workflow files which could contain malicious instructions.
Ingestion points: Workflow files are read via gh api or from local directories in scripts/cron-schedule-planner.sh.
Boundary markers: The generated report provides clear tables and sections, though it does not use explicit prompt injection boundary markers for the ingested text.
Capability inventory: The skill can suggest file modifications, but the instructions require explicit user confirmation before any writes occur. It uses grep, sed, and awk for processing.
Sanitization: Content is processed via regex-based extraction (grep/sed) rather than direct interpolation into agent prompts, reducing the risk of the agent misinterpreting the workflow content as instructions.

cron-schedule-planner