Trust Assessment
zhipu-search received a trust score of 87/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 3 findings: 0 critical, 0 high, 2 medium, and 0 low severity. Key findings include Suspicious import: requests, Prompt Injection via search_query parameter, Potential PII transmission via user_id parameter.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 14, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings3
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Suspicious import: requests Import of 'requests' detected. This module provides network or low-level system access. Verify this import is necessary. Network and system modules in skill code may indicate data exfiltration. | Static | skills/whyhit2005/zhipu-web-search/scripts/zhipu_search.py:11 | |
| MEDIUM | Prompt Injection via search_query parameter The `search_query` parameter is directly incorporated into the `messages` array as user content for the Zhipu LLM (`glm-4-flash`) without sanitization. A malicious user could craft a `search_query` to attempt prompt injection against the Zhipu LLM, potentially manipulating its behavior, extracting system prompts, or generating unexpected responses. While the skill itself primarily extracts data from the Zhipu LLM's response, a successful injection could lead to misleading search results or unintended interactions with the downstream LLM. Implement input sanitization or validation for `search_query` to mitigate prompt injection risks against the downstream LLM. Consider using a dedicated prompt templating library or a content filtering mechanism if the Zhipu API does not offer built-in protections. Alternatively, clearly document this risk to users of the skill. | LLM | scripts/zhipu_search.py:63 | |
| INFO | Potential PII transmission via user_id parameter The skill allows an optional `user_id` parameter to be passed directly to the Zhipu API. While the skill itself does not generate or collect this ID, if the calling agent or user provides personally identifiable information (PII) as the `user_id`, this data will be transmitted to the Zhipu API. This is a privacy consideration, as the skill acts as a conduit for this data. The skill's documentation describes `user_id` as 'End user ID (6-128 chars)', which could imply PII. Document clearly that no PII should be passed in the `user_id` parameter unless explicitly consented to by the end-user and compliant with relevant privacy regulations. Advise users to use anonymized or session-specific identifiers instead of direct PII. Consider adding a warning or validation if the `user_id` format appears to be PII (e.g., email address). | LLM | scripts/zhipu_search.py:90 |
Scan History
Embed Code
[](https://skillshield.io/report/cfbe0b413ed5dc42)
Powered by SkillShield