More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    workspace governance

    1

    Audit, reorganize, and govern your workspace to reduce AI token overhead and stop agents from dumping files in repos.

    Free

    Security scanned
    workspace governance

    workspace governance

    Example session with this skill installed

    Audit my home directory for AI context overhead and suggest a reorganization plan.

    • Read your context and instructions
    • Compiled the workspace governance
    • Generated the document

    I have completed a read-only audit of your home directory using scripts/audit-workspace.py. I identified 14.2k tokens of overhead from legacy .cursorrules and 4.2GB of reclaimable space in node_modules.

    Workspace Refactoring Plan

    Current PathTarget Path (Governed)TypeToken Impact
    ~/Documents/Notes~/knowledge/study/generalSSOT Vault-1.2k
    ~/dev/old-project~/archives/old-projectDehydration-800
    ~/models/llama3~/ai-models/llm/llama3Read-Only Ref0
    ~/scripts/ai-skills~/agent-hub/skillsSSOT Skill-2.1k

    Migration Commands (macOS/Linux)

    
    # 1. Create parent folders only (mv creates the final folder)
    mkdir -p ~/knowledge/study ~/agent-hub
    
    # 2. Move and link: knowledge
    mv ~/Documents/Notes ~/knowledge/study/general
    ln -sfn ~/knowledge/study/general ~/Documents/Notes
    
    # 3. Move and link: skills
    mv ~/scripts/ai-skills ~/agent-hub/skills
    ln -sfn ~/agent-hub/skills ~/scripts/ai-skills
    

    Next steps

    1. Open the full interactive report at ./dist/.cache/dry-run-report.html to review specific file flags.
    2. After you approve, run the mv and ln commands to apply the new structure.
    3. Confirm if you want to proceed with "dehydrating" ~/archives/ to recover 4.2GB of disk space.

    dry-run-report.html

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Stop agents from turning your repositories into junkyards. This skill solves the "CWD dumping" problem where AI assistants scatter temporary scripts, notes, and build artifacts across your active projects. It provides a structured governance framework to keep code, RAG knowledge, and model weights physically separated while maintaining developer access through a zero-downtime migration system.

    What it does

    • Workspace Auditing scans directories for AI-related files to calculate real token overhead and identifies reclaimable disk space from heavy dependency folders.
    • SSOT Enforcement establishes a Single Source of Truth for knowledge, model weights, and skills using a read-only permission matrix.
    • Zero-Downtime Migration moves legacy folders to a governed structure while leaving symlinks or junctions behind to prevent breaking existing scripts or IDE shortcuts.
    • Bilingual Reporting generates interactive HTML dry-run reports in English and Traditional Chinese (zh-TW) with platform-specific command toggles.
    • Spatial Routing automatically diverts non-code outputs like research notes or side-projects to dedicated sandboxes and knowledge vaults.

    How it works

    1. Audit Phase runs the audit-workspace.py script to map your current directory structure and calculate token bills for AI-loaded files.
    2. Review Phase generates an interactive HTML report showing exactly which folders will move and the specific ln or New-Item commands required.
    3. Execution Phase runs only after you approve the dry-run plan: it moves folders and creates symlinks (macOS/Linux) or Directory Junctions (Windows) so existing paths keep working.
    4. Cleanup Phase dehydrates archives by stripping node_modules and .venv from legacy projects after user confirmation.

    Frameworks & tools

    The core logic relies on a standalone Python 3 auditing script. It supports macOS, Linux, and Windows (PowerShell) workflows and integrates with AI coding tools like Claude Code, Cursor, and Windsurf by managing rules like .cursorrules and CLAUDE.md.

    Why this beats prompting it yourself

    Generic prompts often result in destructive file moves that break your build paths. This skill uses a specialized auditor and symlink strategy that ensures your tools can still find files while the agent's context stays lean and focused.

    Use cases

    • Context Reduction to shrink agent token usage by moving non-essential docs out of the active code root.
    • Machine Reorganization to transition a messy home directory into a governed AI-first workspace.
    • Project Scaffolding to link global RAG knowledge or shared skills into new project folders without duplication.
    • Disk Recovery to find and purge deep node_modules or __pycache__ hidden in old archives.

    Safety

    The auditor script is read-only: it never moves, deletes or uploads files, makes no network calls, and uses only the Python 3 standard library. Migration commands are shown for you to review, and nothing moves without your approval. Run it with --demo to see a sample report without scanning your machine.

    Known limitations

    Requires Python 3 for the auditing script. Windows users must use PowerShell for Junction creation. The HTML report requires an internet connection to load Tailwind CSS via CDN for styling.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.

    Requires Python 3. Windows moves require PowerShell. HTML report styles require CDN access. Script does not execute moves automatically; user must run generated commands.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed today

    • Free with an Agensi account

    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listedtoday

    What's inside

    Frequently Asked Questions

    Its routing rules tell the agent where side-quest outputs belong: notes go to ~/knowledge/, new tools and side projects to ~/projects/ or ~/agent-lab/, global plugins and skills to ~/agent-hub/. The current repo only receives work that belongs to it.

    It is a standard SKILL.md skill, so it works with Claude Code, Cursor, Codex, Copilot and other agents that load skills. The auditor also measures context files from these tools, including CLAUDE.md, AGENTS.md, .cursorrules and .github/copilot-instructions.md.

    The migration process uses symlinks (macOS/Linux) and Directory Junctions (Windows), ensuring your IDE shortcuts, scripts, and build paths continue to work without any manual reconfiguration.

    The SKILL.md governance rules and migration SOP, a read-only Python 3 auditor script (scripts/audit-workspace.py), and a sample bilingual dry-run report (assets/demo-report.html).

    It supports macOS, Linux and Windows. On Windows it uses PowerShell Directory Junctions, which do not require administrator rights.

    The auditor itemizes the token cost of every auto-loaded AI file (CLAUDE.md, AGENTS.md, .cursorrules, SKILL.md and others), so you can see which files bloat every session's context. Separately, it measures reclaimable disk space in folders such as node_modules and .venv.

    New in Agents & Orchestration

    Free