Back to skill

Security audit

Openclaw Cc Rules

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed programming-workflow rule set that guides planning, read-only exploration, task tracking, and conservative Git use rather than adding hidden code or unsafe behavior.

Install this if you want OpenClaw to apply a conservative coding workflow across programming tasks. Review the broad activation scope and language preference first, and customize or disable rules if you only want them for specific projects or sessions.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (13)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The README states that any programming-related content will automatically invoke the skill, creating an overly broad activation scope. Broad triggers increase the chance the skill will activate in unintended contexts, causing hidden behavioral override, prompt hijacking of normal sessions, or unreviewed application of persistent workflow rules.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
70% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · README.md (reported line 78)May include surrounding context.

md
### 🛡️ Git 安全协议 — 不作死

**绝对禁止(除非你明确要求):**
- `git push --force`
- `git reset --hard`
- `git checkout .` / `git restore .`
- `--no-verify` / `--no-gpg-sign`

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
70% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 156)May include surrounding context.

md
### 🛡️ Git 安全协议 — 不作死

**绝对禁止(除非你明确要求):**
- `git push --force`
- `git reset --hard`
- `git checkout .` / `git restore .`
- `--no-verify` / `--no-gpg-sign`

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
65% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · README.md (reported line 79)May include surrounding context.

md
**绝对禁止(除非你明确要求):**
- `git push --force`
- `git reset --hard`
- `git checkout .` / `git restore .`
- `--no-verify` / `--no-gpg-sign`
- `git commit --amend`(除非你说"修改上次 commit")

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
65% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 157)May include surrounding context.

md
**绝对禁止(除非你明确要求):**
- `git push --force`
- `git reset --hard`
- `git checkout .` / `git restore .`
- `--no-verify` / `--no-gpg-sign`
- `git commit --amend`(除非你说"修改上次 commit")

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 175)May include surrounding context.

md
2. 分析变更:
   - 概括改动性质(新功能 / 增强 / 修复 / 重构 / 测试 / 文档)
   - 排除含敏感信息的文件(.env、credentials.json 等)
   - 写 1-2 句 commit message,聚焦"为什么改"而不是"改了什么"

3. 执行提交:

Session Persistence

Medium
Category
Rogue Agent
Confidence
91% confidence
Finding

This finding is substantively the same persistence issue: the README instructs users to copy the skill into a durable local skills directory so it remains active across sessions. In the context of agent skills, persistent behavioral modification without strict activation boundaries increases the risk of unintended influence over future interactions.

Content

Scanner excerpt · README.md (reported line 196)May include surrounding context.

bash
# 1. 创建 skill 目录
mkdir -p ~/.openclaw/skills/cc-rules

# 2. 复制 SKILL.md
cp SKILL.md ~/.openclaw/skills/cc-rules/

Session Persistence

Medium
Category
Rogue Agent
Confidence
91% confidence
Finding

This finding is substantively the same persistence issue: the README instructs users to copy the skill into a durable local skills directory so it remains active across sessions. In the context of agent skills, persistent behavioral modification without strict activation boundaries increases the risk of unintended influence over future interactions.

Content

Scanner excerpt · README.md (reported line 196)May include surrounding context.

bash
# 1. 创建 skill 目录
mkdir -p ~/.openclaw/skills/cc-rules

# 2. 复制 SKILL.md
cp SKILL.md ~/.openclaw/skills/cc-rules/

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

Using a common phrase like asking the assistant to review a project's architecture as a verification trigger is risky because it overlaps with ordinary conversation. This makes accidental activation more likely and weakens user awareness and control over when the skill's instructions take effect.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · README.md (reported line 223)May include surrounding context.

md
## 自定义

你可以编辑 `~/.openclaw/skills/cc-rules/SKILL.md` 来调整规则:

### 调整 Plan Mode 触发条件

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The activation conditions are extremely broad and cover most software-development interactions, causing the skill to auto-apply to a wide range of unrelated requests. This can override user intent, inject process constraints into ordinary conversations, and increase the chance that risky tool-use guidance is surfaced when not needed.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

Line L260 says the skill 'does not include specific tools or API calls', which implies a purely normative workflow skill. However, the README repeatedly prescribes concrete command usage such as ls/cat/grep/find/git and even clone/install steps, so the documentation understates the operational behavior it is instructing the agent to perform.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

All user-facing instructions in this file are presented in Chinese, and the skill does not state that language selection is optional or user-driven. This can violate a language-choice policy when the skill is applied in broader contexts without explicit user opt-in.

Content

No source excerpt is available for this finding.