ClawHub
Back to skill

Security audit

B2B供应商筛选专家

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed B2B factory-search helper that sends the user's manufacturer-search terms to a documented external search API and returns public supplier profile information.

Install this only if you are comfortable sending factory-search queries and filters to the documented external B2B search service. Use it for manufacturer sourcing, not purchasing or price comparison, and keep host permissions limited to running the bundled lookup script.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill instructs the agent to execute a shell script (`bash scripts/search.sh ...`) but does not declare any explicit tool scope such as allowed tools or permissions. This creates a governance gap: an orchestrator may permit shell access more broadly than intended, increasing the chance of unintended command execution or misuse if the skill is invoked in the wrong context.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger list includes broad natural-language phrases like '找厂家' and '谁家能代工', which can match many ordinary conversations and cause the skill to activate unexpectedly. Over-broad activation is dangerous because this skill has shell-backed behavior; unintended invocation could launch external queries, consume resources, or steer a conversation into the wrong tool path without clear user intent.

Natural-Language Policy Violations

Medium
Confidence
83% confidence
Finding
The file is entirely written as mandatory behavioral policy for the shared skill set and does not offer any user choice of language or locale. Because this shared guidance governs multiple skills, it implies a Chinese-language operating constraint without explicit opt-in or a documented justification that the skills are region- or language-specific.

External Transmission

Medium
Category
Data Exfiltration
Content
## 联调示例

```bash
curl -s -X POST "https://gwgp-zmtc8kkxxzw.i.bdcloudapi.com/skill_api/factory_search" \
  -H "Content-Type: application/json" \
  -H "X-App-Source: workbuddy" \
  -d '{"query":"铝合金压铸件 定制加工","location":"江苏苏州","page_size":10}'
Confidence
60% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
[ -n "$WITH_PRODUCTS" ] && BODY="$BODY,\"with_products\":$WITH_PRODUCTS"
BODY="$BODY}"

RESP="$(curl -sS --max-time "$TIMEOUT" -X POST "$BASE_URL/skill_api/factory_search" \
    -H "Content-Type: application/json" \
    -H "X-App-Source: $APP_SOURCE" \
    -d "$BODY" 2>&1)" || {
Confidence
70% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Vague Triggers

Low
Confidence
82% confidence
Finding
The routing guidance says to hand off some requests to a product-search skill, but the boundaries remain somewhat subjective ('买现货、比价格' vs. manufacturer search), which can lead to inconsistent or incorrect skill selection. In this case the danger is mainly operational: misrouting can trigger shell-based lookups unnecessarily, return irrelevant results, or bypass the better-scoped skill for the user's actual task.

Static analysis

No suspicious patterns detected.