- Home
- Skills
- Code Quality & Review
- agent-rongchi
Agent Rongchi
A compact behavioral discipline skill (八荣八耻) for coding agents: query before guessing, confirm before executing, test before claiming success.
Free
Works with the AI Agents you already use
agent-rongchi
Example session with this skill installed
I need to add a new user notification service to this repo. Apply coding discipline rules and tell me how you will proceed.
- Read your context and instructions
- Compiled the agent-rongchi
I have applied the coding discipline framework. Before writing any code, I will follow the protocol to ensure architectural compliance and reuse.
Implementation Plan
| Step | Discipline Action | Objective |
|---|---|---|
| 1. Query | grep and ls existing services | Identify established patterns for service instantiation and dependency injection. |
| 2. Reuse | Check src/common/notifications | Find existing notification interfaces or helpers to prevent redundant code. |
| 3. Confirm | Prompt for logic | Ask if notifications should be synchronous or queued via Redis/SQS based on business needs. |
| 4. Validate | Run npm test | Create a minimal test case for the new service before claiming the task is complete. |
Discipline Checkpoints
- Interface Integrity: I will not invent a new
NotificationManagerif aMessagingServicealready exists.
Honest Uncertainty: If I cannot find the SMTP credentials or ENV mapping, I will ask for them rather than guessing the key names.
Minimal Refactor: I will only touch the files necessary for this service, preserving the surrounding architecture.
Next steps
- I will now search the
src/directory for existing notification or messaging patterns. - I will provide a proposed interface for the new service for your review.
- Once approved, I will implement the service and run a local verification test.
Connects securely to your tools. The creator never sees your data.
About this skill
⭐ If this skill helps you, please star us on GitHub: https://github.com/cercatrovato/agent-rongchi
Agent Rongchi — a compact behavioral discipline skill (八荣八耻) for coding agents. One SKILL.md works for Codex, Claude, and other agents: query before guessing interfaces, confirm before ambiguous execution, reuse before creating new interfaces, test before claiming success, respect architecture, admit uncertainty. Includes Chinese, English, and Japanese docs.
Requires the agent to have tool-use capabilities to perform the mandated queries and tests. Does not provide specific test runners or linting configurations.
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
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 month ago
- Free with an Agensi account
Requires
Permissions
No special permissions declared or detected