ClawHub
Back to skill

Security audit

采购拆单优化

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently uses a disclosed external service for procurement planning, with explicit controls for uploads, charges, and project changes.

Confirm that the external provider may receive your supplier quotes and procurement details and retain private projects for 30 days. Prefer a product-specific key for isolation, and verify the current price before approving a paid solve. Provider privacy and retention statements are documented claims, not independently verified guarantees.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

所给代码仅执行配置修改和网关重启，没有读取采购数据、处理供应商报价或计算拆单分配。这些操作也未体现与所声明采购规划功能的支持关系，因此其实际行为与声明用途存在实质性不符。代码中未显示对 API 密钥的读取或外传。

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 · references/HTTP-REQUESTS.md (reported line 21)May include surrounding context.

md
API_ROOT="${AI_SKILLS_API_URL:-https://ai-skills.open-idea.net}"
API_ROOT="${API_ROOT%/}"
# INSPECT_KEY 是本次操作生成并保存的 UUID；重试复用，不能每次重生。
curl --silent --show-error "$API_ROOT/api/v1/procurement-optimizer/purchase.inspect" \
  -H "Authorization: Bearer ${PROCUREMENT_OPTIMIZER_API_KEY:-${AI_SKILLS_API_KEY}}" \
  -H 'Content-Type: application/json' -H "Idempotency-Key: $INSPECT_KEY" \
  --data-binary @inspect.json