The Agent Skills Directory

[SAFE]: The skill does not contain any malicious code, obfuscated content, or unauthorized network operations. Its behavior is consistent with its stated purpose of repo organization.\n- [COMMAND_EXECUTION]: The skill utilizes Python's shutil and pathlib modules, as well as git commands, to perform directory auditing and file moves. These capabilities are used safely within a controlled workflow that includes retries for cloud-synced files and mandatory user approval.\n- [PROMPT_INJECTION]: The skill identifies orphan files by parsing repository content (e.g., README.md, scripts), which presents an indirect prompt injection surface.\n
Ingestion points: Local files and directory structures under audit.\n
Boundary markers: The skill writes a plan to a specific local file path (~/.claude/plans/) and requires explicit confirmation via AskUserQuestion before proceeding with any operations.\n
Capability inventory: Performing file moves, renames, git index removals, and local file writes.\n
Sanitization: The process relies on human-in-the-loop validation of the generated plan to mitigate risks from potentially malicious data in analyzed files.\n- [DATA_EXFILTRATION]: The skill includes logic to detect sensitive file patterns, such as .env, .pem, and token.json, specifically to ensure they are untracked and added to .gitignore. This is a defensive security measure rather than a data exfiltration risk.

organize