Works with the AI Agents you already use
Claude Sentry Skill — Claude 哨兵
A lightweight reminder layer for Claude: your normal answer always comes first, with a short heads-up only when a clear, observable signal appears. No scoring, no ban prediction, no blocking.
Free
Claude Sentry Skill — Claude 哨兵
Example session with this skill installed
Please explain, in plain language, how to back up a small project folder before I make changes, and list the steps in order.
- Read your context and instructions
- Compiled the claude sentry
Here is your normal answer first
- Make a copy of the whole project folder and give it a clear date in its name, for example project-backup-2026-11-02.
- Check that the copy opens and contains the files you expect, especially any configuration files and recent work.
- Store that copy somewhere separate from the folder you are about to change, so a mistake in the working folder does not also affect the backup.
- Only then make your changes in the original folder, one step at a time, and check the result after each step.
- If something goes wrong, you can restore the files from the dated copy and try again.
Short heads-up: you mentioned making changes right after the backup, so this is a good point to confirm the backup copy is complete before you start editing. This reminder does not change, rewrite or block the answer above — it is only an extra note, shown because a clear, observable signal (editing after a backup) appeared in the request.
No score, grade or ban prediction is included, and nothing above was blocked or rewritten.
Connects securely to your tools. The creator never sees your data.
About this skill
Claude Sentry Skill (Claude 哨兵) — version 0.9.1, MIT licensed — is a lightweight reminder layer for Claude.
How it works: your normal answer always comes first. Only when a clear, observable signal appears does the skill add a short heads-up. It does not score or grade you, does not predict bans, does not rewrite your answer, and does not block it.
What is included: a standalone skill package with SKILL.md, reference JSON files, LICENSE and SHA256SUMS. The hooks use Python standard library only — no extra dependencies.
Source: GitHub repo alvin20062006-beep/Claude-Sentry-Skill.
Who it is for: Claude users who want a small, predictable reminder — not a filter, grader or gatekeeper — alongside their normal workflow.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
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 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 15 days ago
- Free with an Agensi account
Permissions
No special permissions declared or detected