The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard development commands such as git, npm test, cargo test, pytest, and go test. These operations are strictly aligned with the skill's stated purpose of managing the development lifecycle and do not involve suspicious or unauthorized command patterns.
[EXTERNAL_DOWNLOADS]: The skill performs standard network operations for development workflows, including git pull, git push, and creating pull requests via the GitHub CLI (gh). These operations target well-known, trusted services necessary for version control.
[PROMPT_INJECTION]: The skill includes an 'EXECUTE PLAN' mode that involves reading and following tasks from a plan file. While this represents a surface for indirect prompt injection if the plan content is untrusted, the skill explicitly instructs the agent to review the plan critically and raise concerns with a human partner before execution, mitigating the risk.
[DATA_EXFILTRATION]: The skill mentions copying .env files when setting up environment workspaces. This is documented as a standard local configuration step for development and does not include any instructions or patterns to transmit sensitive data to external or untrusted domains.

dev-workflow