ClawHub
Back to skill

Security audit

梓享双擎 AI 全网搜索平台（按搜索次数收费）

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed paid web-search connector that sends user searches to its stated third-party API using a user-provided API key.

Install only if you are comfortable sending search queries and domain filters to the stated third-party paid search endpoint and using a ZIXIANGAI_API_KEY for billing. Avoid placing secrets or private internal terms in search queries, and prefer reviewed or fixed versions if you rely on domain filters for policy enforcement.

A.I.G

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 (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/search.sh:55
Finding

Insufficient JSON Validation Allows Search API Request Manipulation

Content
View full analysis
Remediation
View remediation
/dev/null; then echo '{"code":400,"msg":"include_domains must be an array of at most 20 domain strings"}' exit 1 fi if ! jq -e 'type == "array" and length <= 20 and all(.[]; type == "string")' \ <<<"$EXCLUDE_DOMAINS" >/dev/null; then echo '{"code":400,"msg":"exclude_domains must be an array of at most 20 domain strings"}' exit 1 fi PAYLOAD="$( jq -n \ --arg engine "$ENGINE" \ --arg query "$QUERY" \ --argjson max_results "$MAX_RESULTS" \ --arg time_filter "$TIME_FILTER" \ --argjson include_domains "$INCLUDE_DOMAINS" \ --argjson exclude_domains "$EXCLUDE_DOMAINS" \ '{ engine: $engine, query: $query, max_results: $max_results, time_filter: $time_filter, inc ...[truncated 193 chars]

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
- "scripts/search.sh"

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill references a shell script requirement (scripts/search.sh) but does not declare any explicit tool scope such as allowed shell access. Undeclared shell capability increases risk because the runtime may execute commands without clear user/admin visibility into what tools the skill is permitted to use, which can enable command execution, data access, or unintended network activity beyond the documented search API call.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The manifest description is entirely in Chinese and presents the skill as the default interface language, with no indication that users may choose another language. Under the policy, forcing a specific language without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The natural-language instructions and usage documentation are entirely in Chinese, which can force a specific language on users of the skill. The policy allows locale constraints only when explicitly justified or when users are offered a choice, neither of which is present here.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The script transmits user-supplied search queries, optional domain filters, and a bearer API credential to an external third-party service. In the context of an agent skill, this is security-relevant because users or calling systems may unknowingly send sensitive prompts, internal data, or proprietary research terms off-platform, and the script provides no consent gate, data classification check, or safeguard against sensitive-data exfiltration.

Content

Scanner excerpt · scripts/search.sh (reported line 85)May include surrounding context.

sh
"$EXCLUDE_DOMAINS")

# 发送 POST 请求（适配官方接口参数格式）
curl -s -X POST "$API_URL" \
     -H "Content-Type: application/json; charset=utf-8" \
     -H "Authorization: Bearer ${API_KEY}" \
     -d "$PAYLOAD"

Static analysis

No suspicious patterns detected.