Ae1
High
- Category
- analysis-evasion
- Content
- `references/international-search.md` - International search guide
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Security audit
免Api聚合搜索Multi Search Engine
Security checks for vulnerabilities and agentic risk
Overview
This skill is a straightforward search-engine URL guide, with expected outbound searches but no hidden code, persistence, credential use, or local system changes.
Installers should understand that using this skill sends search terms to the selected external search provider. Do not search for passwords, tokens, confidential company material, or personal data unless you are comfortable disclosing those terms to that provider, and treat cached or archived-result examples as something to use only when authorized and appropriate.
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
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (8)
Missing User Warnings
Medium
- Confidence
- 94% confidence
- Finding
- The documentation provides multiple ready-to-use examples that send raw search terms directly to third-party search engines, but it does not warn users that those queries may contain sensitive data and will be disclosed to external providers. In a skill explicitly designed to route user queries to many different engines, including privacy-sensitive and jurisdictionally distinct services, this omission can lead to unintentional leakage of confidential prompts, internal project names, credentials, or personal data.
Natural-Language Policy Violations
Medium
- Confidence
- 94% confidence
- Finding
- The config includes multiple engines explicitly labeled or scoped to the China locale, such as region "cn" and a CN-specific Bing endpoint with ensearch toggles. Because the file provides only fixed locale-specific options and no natural-language indication of user choice or opt-in, it creates a locale policy concern under the rule for forced language/region behavior.
Natural-Language Policy Violations
Medium
- Confidence
- 91% confidence
- Finding
- The file presents all instructions and examples exclusively in Chinese, with no indication that the user can choose another language or that the content is intentionally limited to a Chinese-speaking audience. Under the policy, forcing a specific language without user opt-in is a natural-language locale violation.
Missing User Warnings
Medium
- Confidence
- 94% confidence
- Finding
- The document repeatedly instructs use of `web_fetch` against third-party search engines but does not warn that user queries, metadata, and potentially sensitive research terms will be transmitted to external services. In an agent skill context, readers may copy these examples directly into automated workflows, increasing the risk of unintended privacy leakage and outbound requests to untrusted destinations.
Natural-Language Policy Violations
Low
- Confidence
- 81% confidence
- Finding
- The natural-language content describing the skill is written in Chinese and does not indicate that another language is available or that the user can opt into this locale. This can conflict with language/locale policy when a skill imposes a specific language without user choice.
Missing User Warnings
Low
- Confidence
- 87% confidence
- Finding
- This markdown file includes usage examples calling `web_fetch` against external search engines, which transmits the user's query over the network. The surrounding description does not provide any warning about privacy implications, query disclosure to third-party search providers, or network access behavior.
Missing User Warnings
Low
- Confidence
- 91% confidence
- Finding
- The cached-page example explicitly promotes viewing deleted content via Google cache without any caution about privacy, authorization, or policy concerns. While not an exploit by itself, it normalizes retrieval of content the original publisher may have removed and could encourage questionable or noncompliant investigation behavior.
Static analysis
No suspicious patterns detected.