Todolist Md Clawdbot Copy

Read, summarize, propose edits, and write back changes to Markdown todo files using line-stable bot markers without altering task identity or completing tasks.

This skill contains functioning Drive integration code but the package metadata claims no required credentials or config paths — that mismatch is a red flag. Before installing or running: 1) Inspect and confirm you trust the code; it's safe-looking but will store OAuth refresh tokens and read secret env files by default under /root/clawd/.secrets. 2) Do not run it on a machine with sensitive root secrets; prefer an isolated container or VM. 3) Supply the minimal credential possible (short-lived ACCESS_TOKEN) instead of giving CLIENT_ID/CLIENT_SECRET/REFRESH_TOKEN if you can. 4) If you must use managed OAuth, change the default refresh token path to a directory you control and ensure file permissions are restrictive. 5) Be aware the scripts call sudo and expect a gog CLI — verify the gog binary path and that using sudo -u ubuntu is acceptable in your environment. 6) Ask the skill author to update registry metadata to list required env vars/config paths and to document exactly where tokens are written and how to opt out of persistent storage. If you cannot confirm these fixes, treat the skill as risky and run only in an isolated environment.

Type: OpenClaw Skill Name: Developer: Version: Description: OpenClaw Agent Skill The skill bundle is classified as suspicious due to the explicit use of `sudo -u ubuntu -H env ... gog ...` for external command execution in `scripts/todolist_drive_folder_agent.mjs` and `scripts/todolist_review_drive.py`. While this capability is presented as necessary for interacting with Google Drive via the `gog` CLI, it grants the AI agent the ability to execute arbitrary commands as the `ubuntu` user. This creates a significant Remote Code Execution (RCE) vulnerability, as a malicious prompt could potentially trick the agent into constructing and executing harmful commands, even if the arguments are passed as an array to `execFileSync`/`subprocess.check_output`. This high-risk capability, despite the benign stated purpose, elevates the classification to suspicious rather than benign.