The Agent Skills Directory

[COMMAND_EXECUTION]: Executes multiple GitHub CLI (gh) commands to automate repository management.
Uses gh api user to retrieve the current user login.
Uses gh api repos/.../collaborators to add and verify repository members.
[PRIVILEGE_ESCALATION]: Performs identity switching and permission modification as part of its core functionality.
The skill uses gh auth switch --user to change the active GitHub account context, which allows the agent to act as a different user.
It grants push (write) permissions to a specified account, which is a sensitive administrative action.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from an external project file, creating a vulnerability surface.
Ingestion points: Reads bot_account and owner_account values from the ## Automation config section of CLAUDE.md.
Boundary markers: No explicit boundary markers or warnings are used to isolate the configuration values from the command execution flow.
Capability inventory: The skill possesses the capability to switch authentication contexts and modify repository permissions via the gh CLI.
Sanitization: There is no evidence of validation or sanitization for the account names retrieved from CLAUDE.md before they are interpolated into shell commands.

add-bot