Back to skill

Security audit

DEX代币分析

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to be a DEX token analysis guide, but it asks for broad command execution and includes under-scoped API, export, and automation behavior that users should review before installing.

Install only if you are comfortable with a finance-analysis skill that may use external market/blockchain APIs, environment-provided API keys, local exports, and broad agent command execution. Prefer requiring explicit approval for every command, export, API-key use, and recurring monitoring task, or ask the publisher to remove exec and the backup file scope creep.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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)

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
SKILL.md:28
Finding

Unnecessary Unrestricted Command-Execution Permission

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (11)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The activation criteria are broad and ambiguous, causing the skill to trigger in loosely related situations rather than only when clearly appropriate. Over-broad activation can route unrelated user content into a high-permission skill that discusses APIs and exec, increasing the chance of unintended data exposure or risky tool use.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

Declaring command execution for a DEX token analysis skill expands the attack surface far beyond what is necessary for a documentation-driven analysis workflow. If an agent enables exec based on this manifest, later prompts or skill logic could invoke local commands, access files, or misuse system resources under the guise of market analysis.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The documentation describes use of external cryptocurrency data sources but does not clearly disclose that prompts, token addresses, or other supplied data may be transmitted to third-party services. This creates a privacy and compliance risk, especially in enterprise environments where users may assume analysis remains local.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill references API keys and command execution without an explicit warning about privacy, credential handling, or system-impacting behavior. Users may provide secrets or approve usage without understanding that the skill could interact with external services or execute commands in the local environment.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill claims it is a pure/safer SKILL.md-style artifact with risky code removed, yet it simultaneously advertises executable capability. This mismatch can mislead users and orchestrators into granting elevated trust or permissions they would not otherwise allow, increasing the chance of unexpected command execution.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The document claims risk code and external dependencies were removed, yet later requires exec support, external APIs, network access, and API keys. These contradictory trust-building claims can mislead reviewers and users into underestimating the skill's real attack surface, especially where execution and secret handling are involved.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The documentation advertises exporting results to local files without clearly warning that the skill may write to the filesystem. In agent environments, undisclosed file-write behavior can lead to unanticipated persistence of sensitive financial data or modification of local state beyond user expectations.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill appears to encourage broad natural-language invocation without tightly defined trigger boundaries. Ambiguous invocation increases the risk that an agent may activate the skill in contexts not intended by the user, potentially causing unintended network calls, analysis actions, or follow-on execution behavior.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill documentation materially shifts scope from a DEX token analysis tool into a broader stock, portfolio, alerting, export, and automation skill. This kind of capability confusion is dangerous because users and agent platforms may grant permissions or invoke the skill under false assumptions, increasing the chance of unintended actions or misuse of broader data-access and automation features.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill presents scheduled automatic execution as a feature without warning about persistent or recurring actions. In an agent context, undisclosed automation can create ongoing external requests, repeated data processing, alerting, or other side effects after the initial interaction, which materially increases operational and security risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

The description explicitly states that the skill supports Chinese interaction, presenting a language constraint without any user opt-in or alternative locale choice. Under the language/locale policy, forcing a specific language is a violation unless the constraint is optional or clearly justified as region-specific.

Content

No source excerpt is available for this finding.