The Agent Skills Directory

[SAFE]: The skill is designed to manage project tasks across multiple AI sessions using a deterministic event-sourcing model. No malicious patterns or security vulnerabilities were detected.
[COMMAND_EXECUTION]: The skill executes local bash scripts for mission management. All user-provided inputs, such as mission slugs, session IDs, and task IDs, are validated against strict regular expressions (e.g., ^[a-z0-9]([a-z0-9-]*[a-z0-9])?$) to prevent path traversal or shell injection.
[COMMAND_EXECUTION]: The scripts utilize jq for JSON processing, specifically using the --arg and --argjson flags to safely interpolate user data (such as task titles and descriptions) into JSON structures, effectively preventing injection attacks.
[DATA_EXFILTRATION]: No network operations or external data transmissions were found in any of the scripts. All data operations are localized to the docs/ and scripts/ directories.

mission