skills/freezing-edits/SKILL.md
Blocks all file writes and edits outside a specified directory for the rest of the session. Use when debugging, investigating, or doing exploratory work where you want to prevent accidental modifications to unrelated code. Pass the allowed directory as an argument, e.g. /freezing-edits src/components
npx skillsauth add riccardogrin/skills freezing-editsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Locks down file modifications to a single directory for the rest of the session. Everything outside that directory becomes read-only to Claude.
/freezing-edits src/components
This allows edits only to files under src/components/ and blocks Write/Edit to anything else.
When invoked, save the allowed directory to a config file so the hook can read it:
${CLAUDE_PLUGIN_DATA:-$HOME/.claude/data/freezing-edits}/{"allowed_dir": "/absolute/path/to/directory"}The config file path is: ${CLAUDE_PLUGIN_DATA:-$HOME/.claude/data/freezing-edits}/freeze-config.json
If no argument is provided, ask the user which directory to allow.
The skill registers a PreToolUse hook on Write and Edit tools. Before any file modification, the hook checks whether the target file path starts with the allowed directory. If it doesn't, the edit is blocked (exit code 2) and Claude sees which directory is allowed.
The hooks are session-scoped — they last until the session ends.
To remove the freeze within the same session, delete the config file:
rm "${CLAUDE_PLUGIN_DATA:-$HOME/.claude/data/freezing-edits}/freeze-config.json"
Or simply end the session — hooks don't persist.
src/components will also match src/components-old/ if such a directory exists. Use trailing slashes in the config if precision matters.echo > file). It only covers Claude's Write and Edit tools. For full protection, combine with /being-careful.data-ai
Downloads YouTube videos, transcribes audio via OpenAI Whisper, and produces summaries stored locally. Covers yt-dlp download, audio extraction, transcription, caching, and summarization. Use when a YouTube link is shared and the user wants a transcript or summary
development
Runs an adversarial code review loop that spawns independent reviewer and fixer subagents, iterating until only nitpicks remain. Scores findings by confidence, fixes real issues, and re-reviews with fresh eyes — all internal, no GitHub comments. Use when asked to review code, self-review, adversarial review, or polish code before pushing
development
Creates implementation-ready plans through discovery interviews, external research, and codebase analysis. Covers requirements, competitor research, architecture decisions, and change sequencing. Use when planning features, roadmaps, specs, or any work that needs discovery before coding
development
Generates an autonomous game design loop that iteratively expands a game concept into a comprehensive vision and implementation plan across multiple sessions. Covers mechanic exploration, system design, competitor research, and plan generation. Use when developing a game idea from seed concept to full implementation plan