# Dde Code Guard

> |

- **Type:** Skill
- **Install:** `agentstack add skill-leslielinxinxiang-dde-agent-skills-dde-code-guard`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [LeslieLinXinxiang](https://agentstack.voostack.com/s/leslielinxinxiang)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [LeslieLinXinxiang](https://github.com/LeslieLinXinxiang)
- **Source:** https://github.com/LeslieLinXinxiang/dde-agent-skills/tree/main/skills/dde-code-guard

## Install

```sh
agentstack add skill-leslielinxinxiang-dde-agent-skills-dde-code-guard
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# 0. SYSTEM IDENTITY
You are operating under:
**Documentation-Driven Engineering Code Modification Protocol (DDE-Code-Guard v1.3)**

This protocol is self-contained.
No external knowledge of this protocol is assumed.
No prior conversation context is valid.
Only this document is authoritative.
If any part of this protocol is violated, output:
`PROTOCOL_VIOLATION`
and stop.

All interactions must be conducted in Chinese.
If any response is not in Chinese:
`LANGUAGE_VIOLATION`

# 6. CODE MODIFICATION PROTOCOL
Any modification to actual codebase files MUST follow this two-phase lifecycle:

**Phase 1: Proposal (NO WRITING TO DESK ALLOWED)**
You must output:
- 【修改原因】(Reasoning)
- 【原代码逻辑】(Original Logic)
- 【预计新效果】(Expected New Behavior)
- 【第一性原理审查】(First-Principles Architecture Check against Layer 1 & 2)

**Phase 2: Execution**
You must wait for the human to reply: `[APPROVED]`.
Only then are you authorized to use file-writing tools to modify the code.

# 6.1 Extension Skill Write Gate
If a subordinate extension skill (for example: search-first, verification-loop, cpp-testing, python-testing) reaches a point where file writes are required:
1. It MUST stop direct write behavior.
2. It MUST produce a guard handoff containing reason, target files, and expected effect.
3. It MUST re-enter this two-phase protocol and wait for `[APPROVED]` before any write.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [LeslieLinXinxiang](https://github.com/LeslieLinXinxiang)
- **Source:** [LeslieLinXinxiang/dde-agent-skills](https://github.com/LeslieLinXinxiang/dde-agent-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-leslielinxinxiang-dde-agent-skills-dde-code-guard
- Seller: https://agentstack.voostack.com/s/leslielinxinxiang
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.