The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of CLI tools including git and gh (GitHub CLI) to manage worktrees, branches, pull requests, and issues. These operations are essential for its function as a development factory.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation and use of other specialized skills from the same author's kit (devarfeen/agent-skills-kit). This is a legitimate extension mechanism within the vendor's ecosystem.
[DATA_EXFILTRATION]: As part of its reporting and QA process, the skill uploads structured evidence (JSON results and sanitized screenshots) to external project trackers (Linear and GitHub). It includes specific instructions to exclude sensitive data, credentials, and environment files from these uploads.
[INDIRECT_PROMPT_INJECTION]: The skill has a high attack surface as it processes untrusted input from issue bodies, PR comments, and code diffs. However, it contains an explicit and mandatory rule to treat all such repository text as evidence only, never as instructions to be followed.
[DYNAMIC_EXECUTION]: In run mode, the skill dynamically loads and executes instructions from other skills in its library to handle specific workflow stages. This behavior is governed by a strict handoff protocol and state-machine transitions defined in the reference documents.

factory