The Agent Skills Directory

[SAFE]: The skill defines a strict branching model and naming conventions (lowercase English, digits, and hyphens), which serves as a safeguard against command injection and prevents the creation of misleading or incompatible branch names.
[SAFE]: It incorporates mandatory preflight checks, such as verifying the repository state and fetching from the remote, ensuring the agent operates on the most current and valid data before executing changes.
[SAFE]: Governance rules are explicitly defined to protect critical branches (e.g., prohibiting direct pushes to 'main' or 'develop'), enforcing best practices like the use of pull requests for code integration.
[SAFE]: The skill includes instructions to sanitize and translate user-provided descriptions into English kebab-case slugs, which reduces the attack surface for indirect prompt injection by ensuring user input is transformed into a controlled format before use in shell commands.
[SAFE]: No malicious patterns, such as obfuscation, credential exfiltration, or unauthorized remote code execution, were detected in the instructions or configuration files.

skillsman-branch