ClawHub
Back to skill

Security audit

volcengine-sdk-generator

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches its stated SDK-generation purpose, but it includes copy-ready SDK configuration examples that disable TLS or certificate verification without clear warnings.

Review this skill before installing if you expect it to answer SSL, proxy, or transport-security questions. Keep HTTPS and certificate verification enabled, and avoid copying examples that set verify_ssl false, setVerifyingSsl(false), setDisableSSL(true), setSchema('http'), or equivalent options except in tightly controlled test scenarios. Also treat user-supplied API Params as data sent to Volcengine API Explorer.

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

Error
Location
references/sdk-integration-java.md:259
Finding

Java SDK Example Disables TLS Certificate Verification and HTTPS

Content
View full analysis

Vulnerability Details

File Location: references/sdk-integration-java.md, lines 259–262
Vulnerability Type: Insecure transport configuration
Risk Level: High

Vulnerable Code:

java
apiClient.setVerifyingSsl(false);
apiClient.setDisableSSL(true);
apiClient.setHttpProxy("http://proxy:8080");
apiClient.setHttpsProxy("https://proxy:8080");

Technical Analysis

The Java SDK reference presents copy-ready configuration that disables two fundamental transport protections:

  • setVerifyingSsl(false) disables server certificate verification, allowing a client to accept certificates from an untrusted or impersonated server.
  • setDisableSSL(true) may downgrade SDK communication from HTTPS to plaintext HTTP.

These settings appear in a general “SSL, Proxy, Timeouts” example without a prominent warning, test-only restriction, or secure alternative. Because SKILL.md instructs the agent to answer Java configuration questions directly from this reference, users may receive and copy this configuration when asking about SSL, proxies, or transport settings.

The proxy declarations are not independently vulnerable, but using an HTTP proxy alongside disabled certificate verification or disabled TLS substantially increases interception risk. SDK requests may carry access-key identifiers, session tokens, signed authorization data, sensitive API parameters, and confidential responses.

Attack Path

  1. A user asks the Skill for Java SDK SSL, proxy, or transport configuration.
  2. Following SKILL.md, the agent uses this integration reference to construct its answer.
  3. The user copies the example, including setVerifyingSsl(false) or setDisableSSL(true).
  4. The SDK communicates over an untrusted network or through a malicious or compromised proxy.
  5. With certificate verification disabled, an attacker presents a spoofed certificate and impersonates the API endpoint; with SSL disab ...[truncated 826 chars]
Remediation
View remediation

Remediation Suggestions

  1. Remove apiClient.setVerifyingSsl(false) and apiClient.setDisableSSL(true) from the standard example.
  2. Make the secure defaults explicit:
    • Keep HTTPS enabled.
    • Keep certificate and hostname verification enabled.
    • Use an HTTPS proxy where appropriate.
  3. For private certificate authorities, document how to configure a restricted custom trust store rather than disabling verification globally.
  4. If the insecure switches must be documented for diagnostics, move them to a clearly labeled test-only section and add warnings that they must never be used with production endpoints, real credentials, or sensitive data.
  5. Add a warning that SDK debug output and proxy infrastructure may expose sensitive headers or request data.
  6. Add documentation tests or security linting to reject examples that disable TLS or certificate verification unless accompanied by an explicit, prominent test-only warning.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (22)

Tp4

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The skill description says Params must come from the user, but the body also authorizes auto-mocking missing required parameters, local swagger caching, and post-processing generated code. This mismatch can mislead reviewers and downstream policy systems about what the skill actually does, causing unsafe trust in outputs and unreviewed data handling behaviors.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill description says Params must come from the user, but the body also authorizes auto-mocking missing required parameters, local swagger caching, and post-processing generated code. This mismatch can mislead reviewers and downstream policy systems about what the skill actually does, causing unsafe trust in outputs and unreviewed data handling behaviors.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The skill description says Params must come from the user, but the body also authorizes auto-mocking missing required parameters, local swagger caching, and post-processing generated code. This mismatch can mislead reviewers and downstream policy systems about what the skill actually does, causing unsafe trust in outputs and unreviewed data handling behaviors.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

md
th `Channel=service` and `Channel=api`. After a concrete API is selected, call `scripts/make_code.py` in direct mode with `--service-code`, `--api-version`, and

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.

md
th `Channel=service` and `Channel=api`. After a concrete API is selected, call `scripts/make_code.py` in direct mode with `--service-code`, `--api-version`, and

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

md
th `Channel=service` and `Channel=api`. After a concrete API is selected, call `scripts/make_code.py` in direct mode with `--service-code`, `--api-version`, and

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 70)May include surrounding context.

md
th `Channel=service` and `Channel=api`. After a concrete API is selected, call `scripts/make_code.py` in direct mode with `--service-code`, `--api-version`, and

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 72)May include surrounding context.

md
th `Channel=service` and `Channel=api`. After a concrete API is selected, call `scripts/make_code.py` in direct mode with `--service-code`, `--api-version`, and

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 92)May include surrounding context.

md
th `Channel=service` and `Channel=api`. After a concrete API is selected, call `scripts/make_code.py` in direct mode with `--service-code`, `--api-version`, and

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 103)May include surrounding context.

md
th `Channel=service` and `Channel=api`. After a concrete API is selected, call `scripts/make_code.py` in direct mode with `--service-code`, `--api-version`, and

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill instructs use of shell, network, and local file operations but does not declare any tool scope or allowed-tools boundary. In an agent environment, this increases the chance of over-privileged execution, making unintended network access, filesystem writes, or shell invocation possible if the runtime grants broad defaults.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The document explicitly shows apiClient.setVerifyingSsl(false) and apiClient.setDisableSSL(true) as normal configuration options without any warning, restriction, or explanation of the security consequences. In an SDK integration reference, this can lead developers to disable certificate validation or TLS entirely in production, enabling man-in-the-middle attacks, credential theft, and tampering with API requests and responses.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The documentation explicitly demonstrates setSchema('http') and setVerifySsl(false) without any warning or restriction to test-only scenarios. In an SDK integration reference, readers may copy these settings into production, which can disable transport security and enable man-in-the-middle interception, credential theft, and response tampering.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The documentation explicitly shows configuration.scheme = "http" and configuration.verify_ssl = False as normal configuration examples without any warning that these settings disable transport security. Consumers may copy-paste this into production code, allowing man-in-the-middle attacks, credential exposure, and response tampering when calling Volcengine APIs.

Content

No source excerpt is available for this finding.

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
99% confidence
Finding

verify_ssl = False disables certificate validation, which defeats TLS authentication even if HTTPS is used and makes interception or endpoint impersonation trivial on untrusted networks. In this skill's context, the SDK handles API requests that may carry AK/SK, session tokens, and signed requests, so weakening TLS protection is especially dangerous.

Content

Scanner excerpt · references/sdk-integration-python.md (reported line 229)May include surrounding context.

python
configuration.scheme = "http"
configuration.verify_ssl = False
configuration.ssl_ca_cert = "/path/to/ca-bundle.crt"
configuration.proxy = "http://proxy:8080"
configuration.http_proxy = "http://proxy:8080"

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The HTTP helper defaults the x-language header to zh, causing upstream API interactions to prefer Chinese unless the caller overrides it. This is reinforced elsewhere by Chinese-only warning text, making the skill impose a locale by default rather than offering a neutral or opt-in choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The generated mock-value notice is always emitted in Chinese, regardless of user preference or execution context. This is a natural-language locale policy issue because the skill forces a specific language for user-facing output without opt-in.

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 · scripts/make_code.py (reported line 22)May include surrounding context.

python
from typing import Any


API_BASE = "https://api.volcengine.com/api/common"
API_VERSION_PATTERN = re.compile(r"\b20\d{2}-\d{2}-\d{2}\b")

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/rg_rank.py (reported line 20)May include surrounding context.

python
from typing import Any


API_BASE = "https://api.volcengine.com/api/common"
API_VERSION_PATTERN = re.compile(r"\b20\d{2}-\d{2}-\d{2}\b")

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/sdk_info.py (reported line 22)May include surrounding context.

python
from typing import Any


API_BASE = "https://api.volcengine.com/api/common"
API_VERSION_PATTERN = re.compile(r"\b20\d{2}-\d{2}-\d{2}\b")

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The request headers hard-code x-language to zh, which imposes a specific locale on all API calls regardless of user preference. This is a natural-language policy issue because the file provides no user choice, opt-in, or documented reason for restricting responses to Chinese.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The instruction says mock comments 'must be in Chinese by default,' which imposes a language choice on generated output regardless of the user's preferred language. This is a natural-language locale policy concern because the file does not offer opt-in or an explicit user language choice for those comments.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.