The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the subprocess module to run Git and GitHub CLI commands.
Evidence: In scripts/repo-rename.py, the run function executes commands like git status, git ls-files, gh repo view, and gh repo rename. These are used to validate the repository state and perform the actual rename operations.
[INDIRECT_PROMPT_INJECTION]: The skill reads and modifies files within the repository and AI agent transcript directories, which represents a potential attack surface for indirect prompt injection.
Ingestion points: scripts/repo-rename.py reads content from files within the repository (using git ls-files or os.walk) and from AI agent project stores (~/.claude/projects, ~/.codex/sessions).
Boundary markers: None; the skill performs literal byte-string replacements on the ingested content without delimiters or safety warnings.
Capability inventory: The skill can execute git and gh commands, rename directories (os.rename), and write to files (Path.write_bytes).
Sanitization: Input for the new repository name is validated to prevent directory traversal (/ or . checks), and paths used for agent continuity are normalized using regex in claude_project_name.

repo-rename