The Agent Skills Directory

[SAFE]: The skill implements a standard software engineering workflow (Commit -> Push -> PR -> Merge -> Close Issue) using legitimate tools (git and gh). No malicious patterns, obfuscation, or unauthorized network access were detected.
[COMMAND_EXECUTION]: The skill uses the Bash(git:*) and Bash(gh:*) tools, which are explicitly listed in the allowed-tools manifest. The commands are scoped to repository management and do not attempt to access sensitive system files or execute arbitrary code.
[PROMPT_INJECTION]: The instructions do not contain any bypass markers, role-play injections, or attempts to override the agent's core safety instructions.
[SAFE]: While the skill interpolates user-provided text into commit messages and PR bodies, the provided examples use quoted heredocs (cat <<'EOF'). This is a defensive programming practice that prevents the shell from interpreting special characters or performing command substitution within the user-provided string.

ship-it