Back to skill

Security audit

ict

Security checks for vulnerabilities and agentic risk

Overview

This is mostly a local security scanner, but it needs review because it can expose detected secrets in reports and makes an undocumented update request.

Review before installing. Use it only in environments where outbound access to api.clawhub.ai is acceptable, avoid running the sensitive-info scanner on repositories with real credentials unless reports are protected and redacted, and do not rely on this tool as the sole approval gate because its exclusion logic can miss malicious files with certain names or paths.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (3)

T09 · Insecure Skill Coding Practices

Error
Location
ict.py:48
Finding

Attacker-Controlled File and Directory Names Can Bypass Security Scanning

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
sensitive_info_scanner.py:176
Finding

Detected Credentials Are Copied into Findings and Plaintext Reports

Content
View full analysis
Remediation
View remediation

other

Note
Location
ict.py:16
Finding

Every CLI Invocation Performs an Undocumented Outbound Update Request

Content
View full analysis
curr: print("\n" + "="*50) print("Update Available!") print(f"Current: v{VERSION}") print(f"Latest: v{latest}") print("Please update: clawhub install ict") print("="*50 + "\n") except: pass except: pass ``` The request occurs before CLI argument handling: ```python def main(): check_for_update() p = argparse.ArgumentParser(description='ICT Skill Auditor v' + VERSION) ``` ### Technical Analysis The primary declared functionality is local static analysis, which does not require network access. Nevertheless, `main()` calls `check_for_update()` before parsing arguments, causing all CLI invocations—including local audits and help requests—to contact `api.clawhub.ai`. The request does not include scanned files, credentials, audit results, or explicit user identifiers. The destination also belongs to the declared ClawHub ecosystem. Therefore, this is not evidence of credential exfiltration or remote payload execution. However, the request reveals normal network metadata such as source IP add ...[truncated 1325 chars]
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
Findings (103)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

Scanning for secrets and recursively reading config files is a meaningfully different and more privacy-sensitive behavior than checking source for dangerous code patterns. That mismatch can lead to unexpected access to credentials and configuration data, increasing exposure if the tool logs, reports, or mishandles findings.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

Scanning for secrets and recursively reading config files is a meaningfully different and more privacy-sensitive behavior than checking source for dangerous code patterns. That mismatch can lead to unexpected access to credentials and configuration data, increasing exposure if the tool logs, reports, or mishandles findings.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Scanning for secrets and recursively reading config files is a meaningfully different and more privacy-sensitive behavior than checking source for dangerous code patterns. That mismatch can lead to unexpected access to credentials and configuration data, increasing exposure if the tool logs, reports, or mishandles findings.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Scanning for secrets and recursively reading config files is a meaningfully different and more privacy-sensitive behavior than checking source for dangerous code patterns. That mismatch can lead to unexpected access to credentials and configuration data, increasing exposure if the tool logs, reports, or mishandles findings.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Scanning for secrets and recursively reading config files is a meaningfully different and more privacy-sensitive behavior than checking source for dangerous code patterns. That mismatch can lead to unexpected access to credentials and configuration data, increasing exposure if the tool logs, reports, or mishandles findings.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Scanning for secrets and recursively reading config files is a meaningfully different and more privacy-sensitive behavior than checking source for dangerous code patterns. That mismatch can lead to unexpected access to credentials and configuration data, increasing exposure if the tool logs, reports, or mishandles findings.

Content

No source excerpt is available for this finding.

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 · SKILL.md (reported line 86)May include surrounding context.

md
| 2 | Code Execution | eval/exec/spawn |
| 3 | Data Exfiltration | webhook.site, requestbin, ngrok URLs |
| 4 | Base64 Obfuscation | Encoded payloads |
| 5 | Sensitive FS | /etc/passwd, ~/.ssh, ~/.aws |
| 6 | Crypto Wallet | ETH/BTC address detection |
| 7 | Dependency Confusion | @internal, typosquatting |
| 8 | Install Hooks | pre/post install |

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 · SKILL.md (reported line 236)May include surrounding context.

md
| 2 | Code Execution | eval/exec/spawn |
| 3 | Data Exfiltration | webhook.site, requestbin, ngrok URLs |
| 4 | Base64 Obfuscation | Encoded payloads |
| 5 | Sensitive FS | /etc/passwd, ~/.ssh, ~/.aws |
| 6 | Crypto Wallet | ETH/BTC address detection |
| 7 | Dependency Confusion | @internal, typosquatting |
| 8 | Install Hooks | pre/post install |

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 · rules.json (reported line 110)May include surrounding context.

json
| 2 | Code Execution | eval/exec/spawn |
| 3 | Data Exfiltration | webhook.site, requestbin, ngrok URLs |
| 4 | Base64 Obfuscation | Encoded payloads |
| 5 | Sensitive FS | /etc/passwd, ~/.ssh, ~/.aws |
| 6 | Crypto Wallet | ETH/BTC address detection |
| 7 | Dependency Confusion | @internal, typosquatting |
| 8 | Install Hooks | pre/post install |

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 · rules.json (reported line 114)May include surrounding context.

json
| 2 | Code Execution | eval/exec/spawn |
| 3 | Data Exfiltration | webhook.site, requestbin, ngrok URLs |
| 4 | Base64 Obfuscation | Encoded payloads |
| 5 | Sensitive FS | /etc/passwd, ~/.ssh, ~/.aws |
| 6 | Crypto Wallet | ETH/BTC address detection |
| 7 | Dependency Confusion | @internal, typosquatting |
| 8 | Install Hooks | pre/post install |

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 · run_tests.py (reported line 104)May include surrounding context.

python
| 2 | Code Execution | eval/exec/spawn |
| 3 | Data Exfiltration | webhook.site, requestbin, ngrok URLs |
| 4 | Base64 Obfuscation | Encoded payloads |
| 5 | Sensitive FS | /etc/passwd, ~/.ssh, ~/.aws |
| 6 | Crypto Wallet | ETH/BTC address detection |
| 7 | Dependency Confusion | @internal, typosquatting |
| 8 | Install Hooks | pre/post install |

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 · SKILL.md (reported line 92)May include surrounding context.

md
| 8 | Install Hooks | pre/post install |
| 9 | Symlink Attack | Symlink to sensitive paths |
| 10 | Time Bomb | Delayed trigger |
| 11 | Remote Exec | curl | bash, wget | sh |
| 12 | Telemetry | Analytics SDK, tracking |
| 13 | Prompt Injection | "ignore previous instructions" |
| 14 | Stealth Exfil | Hidden data transmission |

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 · SKILL.md (reported line 242)May include surrounding context.

md
| 8 | Install Hooks | pre/post install |
| 9 | Symlink Attack | Symlink to sensitive paths |
| 10 | Time Bomb | Delayed trigger |
| 11 | Remote Exec | curl | bash, wget | sh |
| 12 | Telemetry | Analytics SDK, tracking |
| 13 | Prompt Injection | "ignore previous instructions" |
| 14 | Stealth Exfil | Hidden data transmission |

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 · SKILL.md (reported line 92)May include surrounding context.

md
| 8 | Install Hooks | pre/post install |
| 9 | Symlink Attack | Symlink to sensitive paths |
| 10 | Time Bomb | Delayed trigger |
| 11 | Remote Exec | curl | bash, wget | sh |
| 12 | Telemetry | Analytics SDK, tracking |
| 13 | Prompt Injection | "ignore previous instructions" |
| 14 | Stealth Exfil | Hidden data transmission |

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 · SKILL.md (reported line 242)May include surrounding context.

md
| 8 | Install Hooks | pre/post install |
| 9 | Symlink Attack | Symlink to sensitive paths |
| 10 | Time Bomb | Delayed trigger |
| 11 | Remote Exec | curl | bash, wget | sh |
| 12 | Telemetry | Analytics SDK, tracking |
| 13 | Prompt Injection | "ignore previous instructions" |
| 14 | Stealth Exfil | Hidden data transmission |

Instruction Override

High
Category
Prompt Injection
Confidence
80% confidence
Finding

This pattern attempts to override system instructions or ignore safety constraints. Without LLM analysis, manual review is recommended.

Content

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

md
| 10 | Time Bomb | Delayed trigger |
| 11 | Remote Exec | curl | bash, wget | sh |
| 12 | Telemetry | Analytics SDK, tracking |
| 13 | Prompt Injection | "ignore previous instructions" |
| 14 | Stealth Exfil | Hidden data transmission |
| 15 | C2 Server | C2 server detection |
| 16 | Container Escape | Docker socket escape |

YARA rule 'agent_skill_prompt_injection_hidden_instructions': Prompt injection or hidden instructions embedded in AI agent skill text [agent_skills]

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 · SKILL.md (reported line 94)May include surrounding context.

md
Sensitive FS | /etc/passwd, ~/.ssh, ~/.aws |
| 6 | Crypto Wallet | ETH/BTC address detection |
| 7 | Dependency Confusion | @internal, typosquatting |
| 8 | Install Hooks | pre/post install |
| 9 | Symlink Attack | Symlink to sensitive paths |
| 10 | Time Bomb | Delayed trigger |
| 11 | Remote Exec | curl | bash, wget | sh |
| 12 | Telemetry | Analytics SDK, tracking |
| 13 | Prompt Injection | "ignore previous instructions" |
| 14 | Stealth Exfil | Hidden data transmission |
| 15 | C2 Server | C2 server detection |
| 16 | Container Escape | Docker socket escape |
| 17 | SSH Remote | SSH/scp commands |
| 18 | Privilege Escalation | sudo, chmod 777 |
| 19 | Hidden Files | Access to .files |
| 23 | Unusual Ports | 4444, 5555, 1337, etc |

### Supported Languages
- Python (.py)
- Shell (.sh, .bash)
- JavaScript/TypeScript (.js, .ts)

### 5-Dimension Trust Score (0-100)

| Dimension | Max | Description |
|----------|-----|-------------|
| Security | 35 | Security check results |
| Quality

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 · SKILL.md (reported line 99)May include surrounding context.

md
| 15 | C2 Server | C2 server detection |
| 16 | Container Escape | Docker socket escape |
| 17 | SSH Remote | SSH/scp commands |
| 18 | Privilege Escalation | sudo, chmod 777 |
| 19 | Hidden Files | Access to .files |
| 23 | Unusual Ports | 4444, 5555, 1337, etc |

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 · SKILL.md (reported line 249)May include surrounding context.

md
| 15 | C2 Server | C2 server detection |
| 16 | Container Escape | Docker socket escape |
| 17 | SSH Remote | SSH/scp commands |
| 18 | Privilege Escalation | sudo, chmod 777 |
| 19 | Hidden Files | Access to .files |
| 23 | Unusual Ports | 4444, 5555, 1337, etc |

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · rule_loader.py (reported line 35)May include surrounding context.

python
# 转换为与内置规则相同的格式
                rules.append(self._convert_rule_format(rule))
        
        return rules
    
    def _convert_rule_format(self, rule: Dict) -> Tuple:
        """将 JSON 规则转换为内置规则格式 (id, name, patterns, severity)"""

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 · rules.json (reported line 17)May include surrounding context.

json
"\\$API_KEY|\\$SECRET|\\$TOKEN|\\$PASSWORD"
        ],
        "js": [
          "process\\.env",
          "apiKey|api_key|secretKey",
          "\\.env"
        ],

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 · rules.json (reported line 19)May include surrounding context.

json
"\\$API_KEY|\\$SECRET|\\$TOKEN|\\$PASSWORD"
        ],
        "js": [
          "process\\.env",
          "apiKey|api_key|secretKey",
          "\\.env"
        ],

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 · rules.json (reported line 25)May include surrounding context.

json
"\\$API_KEY|\\$SECRET|\\$TOKEN|\\$PASSWORD"
        ],
        "js": [
          "process\\.env",
          "apiKey|api_key|secretKey",
          "\\.env"
        ],

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 · rules.json (reported line 110)May include surrounding context.

json
"severity": "critical",
      "patterns": {
        "py": [
          "/etc/passwd|/etc/shadow",
          "~/.ssh|~/.gnupg|~/.aws"
        ],
        "sh": [

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 · rules.json (reported line 114)May include surrounding context.

json
"severity": "critical",
      "patterns": {
        "py": [
          "/etc/passwd|/etc/shadow",
          "~/.ssh|~/.gnupg|~/.aws"
        ],
        "sh": [