- Home
- Skills
- Agents & Orchestration
- Workspace Toolkit
Works with the AI Agents you already use
Workspace Toolkit
Moving an AI workspace shouldn't mean guessing which folders matter.
$9
Workspace Toolkit
Example session with this skill installed
pack this workspace with the workspace toolkit, exclude secrets, and show me the inspect report first
- Read your context and instructions
- Compiled the workspace toolkit
Inspect report
- Agent directives, settings, skills and workspace docs: included
- Known secret files, tokens and credential folders: excluded by default
- Large reference files: counted, not packed
- Virtual environments: not packed; a freeze file is recorded instead
Next, if you ask to restore
- Impact report — what would change, and where
- Your first confirmation — review the plan
- A local safety snapshot is taken
- Your second confirmation — then write
- A result report: what was restored, what was skipped, what needs a manual step
Nothing is written on the inspect pass. Restore waits for both confirmations.
(This is an illustrative report. A real run uses your actual folders and your actual exclusions.)
Connects securely to your tools. The creator never sees your data.
About this skill
Moving an AI workspace shouldn't mean guessing which folders matter. Or packing your secrets by accident.
Workspace Toolkit is a Python toolkit that packs, inspects and restores an AI agent workspace from one entry point. Secrets are excluded by default. Restore starts with an impact report, a local safety snapshot, and two confirmations before anything is written.
Written by the creator. Directory layout verified on Google Antigravity. Written for Grok Bot migration, with Grok modules and tests included. Other platforms (Claude Code, Cursor, VS Code Cline/Roo) are not yet tested — adapt the folders with your agent before you rely on them.
工作區打包、檢查、還原;預設排除機密;還原前兩次確認。 ワークスペースの梱包・検査・復元。機密は既定で除外。書き込みの前に二重確認。
What's in the package
- The Python toolkit (one entry script, plus modules)
- Interactive trilingual console (English, Traditional Chinese, Japanese — follows system language)
- Pack, inspect and restore modes; optional encryption
- Tests, including Antigravity and Grok end-to-end cases
- Guides in English, Traditional Chinese and Japanese
- Directory layout verified on Google Antigravity; Grok Bot migration is written and tested in-repo, not claimed as a full platform verification
How it works You run the toolkit, or ask your agent to run it. Pack builds a portable archive (split across size-limited parts if needed) and skips known secret files, tokens and credential folders unless you explicitly include them. Inspect shows what would be packed. Restore first writes an impact report, takes a local snapshot, and waits for two confirmations from you before it writes. Oversized reference files are counted, not packed, unless you ask.
Where it stops Restore will not install Python packages for you, and will not write into a platform's live memory store. It stops if a part is missing. It will not overwrite an existing output (it versions the folder). You are responsible for how backup files are stored. Use the built-in encryption when the material is sensitive.
Ready to move a workspace without packing the keys?
Add the skill, then tell your agent: "pack this workspace with the workspace toolkit, exclude secrets, and show me the inspect report first."
This marketplace listing will be kept updated. v2.3.1, one-time purchase. Written by the creator. Directory layout verified on Google Antigravity.
- Directory layout verified on Google Antigravity only. Other platforms are not yet tested.
- Grok Bot migration is written and tested in-repo, not claimed as a full platform verification.
- Restore does not install Python packages and does not write into a platform's live memory store.
- Large reference files are counted, not packed, unless you ask to include them.
- You are responsible for how backup files are stored. Use the built-in encryption when the material is sensitive.
- The listing will be kept updated on this marketplace.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 4 days ago
Requires