Back to skill

Security audit

小红薯短视频全解析

Security checks for vulnerabilities and agentic risk

Overview

This video-analysis skill needs review because it asks for API keys in chat and forces promotional prompts into the workflow.

Configure the key locally instead of pasting it into chat, restrict configuration-file access, and rotate any key already shared. Expect Cue credit charges and mandatory low-balance promotions. Submit only links you permit Cue to process; server retention and the optional Bridge implementation remain unverified. The review verdict rests on credential handling and workflow control, not merely network access or scanner labels.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

The onboarding script requires the agent to solicit the user's complete API key through the conversation, presenting credential disclosure as a necessary setup step.

Medium
Category
Data Exfiltration
Confidence
99% confidence
Finding

The mandatory onboarding script asks users to paste their complete API key into the conversation, exposing a reusable credential to chat processing, history, and any associated retention or logging. The alternative local setup instructions reduce necessity but do not remove the risk because the required script still presents disclosure as the normal setup path; prohibiting subsequent echoes cannot undo the initial exposure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

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

md
CUE_KEY=$(python3 -c "import json,os;p=os.path.expanduser('~/.cue/config.json');print(json.load(open(p)).get('api_key','') if os.path.exists(p) else '')" 2>/dev/null || true)
CUE_KEY=${CUE_KEY:-$CUE_API_KEY}
echo "=== 1/4 API Key ===" && { [ -n "$CUE_KEY" ] && echo "已配置" || echo "未配置!"; }
echo "=== 2/4 远程端点 ===" && curl -sS --max-time 10 -X POST "https://mcp.cuecue.cn/api/omni-reader/mcp/" \
  -H "Authorization: Bearer $CUE_KEY" \
  -H "Content-Type: application/json" -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' \

The skill mandates referral marketing and paid-credit promotion after successful video-processing tasks.

Medium
Category
Excessive Agency
Confidence
90% confidence
Finding

The skill makes promotional messaging mandatory and explicitly prioritizes its fixed referral and purchase prompt over other post-task questions. This is a bounded workflow-hijacking issue: it redirects the assistant toward the provider's commercial goals without user opt-in, although the low-balance condition and absence of automatic purchases substantially limit the impact.

Content

No source excerpt is available for this finding.

Mandatory verbatim Chinese billing prompts impose an output language without user opt-in.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The skill requires the billing question and options to be used verbatim and forbids changing their text, which is entirely Chinese. This prevents adapting these user-facing messages to the user's language, without offering a language choice or explaining why a fixed language is necessary.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 332)May include surrounding context.

解析调用手册

curl 三步

bash
# ① 提交解析(记下返回的 operation_id)