Back to skill

Security audit

Content Publishing Suite

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: the completed scan reported no AIG findings, while the partial Skillspector scan raised a language-policy issue and two subprocess-call alerts.

The verdict is benign, not a guarantee of safety. Before installing, consider whether you are comfortable with a skill that includes executable scripts and a LinkedIn language constraint; the scan was partial, so it did not fully resolve all artifact references or inspect every component.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

The LinkedIn workflow mandates English content without requiring the user's language choice or opt-in.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

The workflow requires a finalized English article before generating any LinkedIn caption, effectively making English mandatory for that channel. Although line 59 describes Chinese or English according to the target audience, this hard prerequisite provides no Chinese-language path or explicit language-selection gate; LinkedIn itself is not an English-only channel.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/validate_publish_output.py (reported line 157)May include surrounding context.

python
command = [sys.executable, args.layout_checker, snippet, "--baseline", args.layout_baseline]
    if title:
        command.extend(["--title", title])
    result = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
    if result.returncode != 0:
        issues.append(("P0", "wechat", "未通过唯一排版基线:标题/小标题/正文字号、行高、颜色、间距或结构发生漂移"))

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/validate_publish_output.py (reported line 221)May include surrounding context.

python
except Exception as e:
        issues.append(("P0", "visual", "摘要卡尺寸读取失败: %s" % e))

    result = subprocess.run(
        [sys.executable, args.summary_checker, args.summary_card_html, "--baseline", args.summary_baseline,
         "--contract", args.summary_contract, "--copy", args.summary_copy],
        stdout=subprocess.PIPE,