Back to skill

Security audit

deepseek-api-search

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed DeepSeek search/fetch helper that sends user queries or URLs to DeepSeek using the user's API key, with no hidden persistence or destructive behavior found.

Install only if you are comfortable sending search queries and fetch URLs to DeepSeek under your DEEPSEEK_API_KEY. Do not use it with private intranet URLs, secret-bearing query strings, or confidential material unless that third-party sharing is acceptable.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

描述覆盖的能力范围明显大于代码实际实现。代码的核心能力是调用 https://api.deepseek.com/v1/tools/search 进行搜索,并可请求搜索结果中附带 content 内容;它并未实现单独的 Fetch API 调用、按 URL 直接读取页面、或站点/语言/日期等专门过滤能力。虽然 need_content 允许搜索结果包含正文,这与“搜索后返回页面正文”部分相关,但不足以支撑“读取指定 URL”“获取网页原文”“Fetch API 抓取”等更明确的声明。因此属于描述与实际行为不符,且是能力缺失型不匹配,而非存在未声明的额外危险能力。

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill explicitly describes network access to DeepSeek APIs and use of the local DEEPSEEK_API_KEY, but it does not declare any tool scope or permissions boundary. This is dangerous because reviewers and policy systems cannot reliably constrain or audit the skill's access to environment secrets and outbound network operations, increasing the chance of unintended data exposure or overbroad execution.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

This script transmits user-supplied URLs to an external DeepSeek API, which can expose sensitive internal URLs, query parameters, or private resources to a third party. In the context of a fetch skill, this is expected functionality, but it still creates a real data-exfiltration and privacy boundary that becomes dangerous if the tool is used on confidential or internal-only targets.

Content

Scanner excerpt · scripts/deepseek_fetch.py (reported line 11)May include surrounding context.

python
from deepseek_search import SearchError as FetchError, request_json

FETCH_URL = "https://api.deepseek.com/v1/tools/fetch"


def validate_url(url: str) -> None:

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code emits validation errors and command descriptions in Chinese, which forces a specific language on users without opt-in. The policy allows locale constraints only when clearly documented and justified, which is not evident here.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The exception message is presented only in Chinese, creating a language policy issue for users who may not understand it. No alternative locale handling or explicit justification for a Chinese-only interface is present in the file.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The parser description, argument help, and runtime error strings are user-visible and hardcoded in Chinese. Because the file does not offer language selection or document a justified locale restriction, this conflicts with the language/locale policy.

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 · SKILL.md (reported line 8)May include surrounding context.

md
from email.utils import parsedate_to_datetime
from typing import Optional

SEARCH_URL = "https://api.deepseek.com/v1/tools/search"


class SearchError(RuntimeError):

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 · scripts/deepseek_search.py (reported line 17)May include surrounding context.

python
from email.utils import parsedate_to_datetime
from typing import Optional

SEARCH_URL = "https://api.deepseek.com/v1/tools/search"


class SearchError(RuntimeError):

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This Python file contains user-facing validation errors, status messages, and CLI descriptions in Chinese only. That creates a language/locale policy issue because the skill forces one language for all users and does not offer any language choice or documented locale constraint.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.