The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git and gh (GitHub CLI) commands to read and write configuration settings, such as git config --global tobico.handle and git symbolic-ref. These are standard operations for development workflow tools.
[INDIRECT_PROMPT_INJECTION]: The skill creates and modifies docs/agents/git-workflow.md, which serves as a set of executable instructions for other agent skills (/to-tasks and /next-task). This represents a vulnerability surface where malicious content in the workflow file could influence future agent behavior.
Ingestion points: Processes data from the existing docs/agents/git-workflow.md file and user-provided inputs via menus.
Boundary markers: No explicit delimiters or boundary markers are used to isolate user-provided text from the instruction templates.
Capability inventory: File system write access to the documentation directory and shell command execution via git and GitHub CLI.
Sanitization: The skill does not perform explicit sanitization or validation on "Free text" inputs provided by the user before writing them into the instruction file.
[DYNAMIC_EXECUTION]: The skill generates shell command sequences based on predefined templates (e.g., for GitHub PR creation, stacked branches, or local merging) and writes them to a file intended to be executed verbatim by other skills.
[EXTERNAL_DOWNLOADS]: The documentation suggests installing the github/gh-stack extension for the GitHub CLI to support stacked pull request workflows.

setup-tobico-skills