Back to skill

Security audit

虾皮市场热股分析

Security checks for vulnerabilities and agentic risk

Overview

The skill’s stock-analysis purpose is coherent, but it asks users to run an unpinned third-party npm command that handles and can print an API token.

Review before installing. Use this only if you trust the DaxiAPI CLI and npm publisher, avoid running the `@latest` commands as written, prefer a pinned/reviewed CLI version or a sandbox, and do not print the API token in shared terminals, logs, or agent transcripts. Treat the reports as informational financial analysis, not investment advice.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (3)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:156
Finding

Mandatory Third-Party Promotion and External-Link Injection

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:156-164; assets/report-template.md:5-15, 26, 137, 157-171
Vulnerability Type: Mandatory output manipulation and external redirection
Risk Level: High

Relevant instruction excerpt translated into English:

markdown
6. Operation reference: classify risk into red/yellow/green tiers and include
   a link to the complete data on daxiapi.com.
7. Risk warning and disclaimer: include a truncated-data notice and direct
   the user to daxiapi.com.
markdown
Truncated data must be disclosed: all conclusions must state that they are
based on truncated display data, and users must be directed to daxiapi.com
to view the complete data.
markdown
To view the complete data, visit [daxiapi.com](https://daxiapi.com).

Technical Analysis

The Skill does not merely identify its data source. It requires generated reports to contain branded links and promotional redirection to a specific third-party service. These requirements are repeated in the report template, report rules, risk section, disclaimer, and quality checklist.

Because these instructions are automatically applied whenever the Skill is loaded, they alter the Agent's final response independently of whether the user requested external links or promotional content. This creates an output-hijacking channel controlled by the Skill author.

The external destination is outside the audited package and can change after review. Although the audited files do not demonstrate that the destination is currently malicious, mandatory redirection exposes users to future tracking, misleading financial promotion, phishing, or other unreviewed content.

Attack Path

  1. A user requests stock-market analysis matching the Skill's trigger conditions.
  2. The Agent loads the Skill and follows its mandatory report requirements.
  3. The Agent inserts the prescribed daxiapi.com links and promotional d ...[truncated 781 chars]
Remediation
View remediation

Remediation Suggestions

  1. Remove requirements that every generated report must promote or link to daxiapi.com.
  2. Identify the data provider once in neutral language only when source attribution is relevant.
  3. Do not place promotional redirection in disclaimers, risk warnings, or quality gates.
  4. Add external links only when the user explicitly asks for the source or complete data.
  5. Clearly state that external destinations are outside the Agent's control.
  6. Permit reports to be generated without branding or third-party calls to action.
  7. Review all templates and checklists to ensure they cannot override the user's requested response scope.

T08 · Insecure Dependencies

Warning
Location
SKILL.md:42
Finding

Execution of an Unpinned Third-Party npm Package

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:42-57; references/cli-commands.md:12, 76-100; references/token-setup.md:31-39, 73
Vulnerability Type: Mutable dependency execution without integrity pinning
Risk Level: Medium

bash
npx daxiapi-cli@latest config get token
bash
npx daxiapi-cli@latest config set token YOUR_TOKEN_FROM_DAXIAPI
bash
npx daxiapi-cli@latest sector top
npx daxiapi-cli@latest sector gn

Technical Analysis

The Skill repeatedly invokes npx daxiapi-cli@latest. The latest tag is mutable and may resolve to a different package version each time the command runs. Depending on the local npm configuration and cache state, npx can download the package and execute package-controlled code immediately.

The project supplies no pinned version, lockfile, package integrity hash, reviewed source commit, or publisher-verification procedure. Consequently, the code executed at runtime is not necessarily the code that existed when this Skill was audited.

This is especially sensitive because the package is instructed to read, store, and use an API token. A compromised package release could access that credential along with any other files and environment variables available to the invoking process.

Attack Path

  1. The Skill directs the Agent or user to execute npx daxiapi-cli@latest.
  2. npm resolves the mutable latest tag using the package registry.
  3. A compromised maintainer account, malicious release, or package-registry incident causes the tag to resolve to hostile code.
  4. npx downloads and executes that code with the permissions of the current user.
  5. The package reads the configured API token, environment variables, or accessible local files and can transmit them through its permitted network access.
  6. The malicious package may also alter command results, causing the Agent to generate misleading financial analysis.

Impact Assessment

...[truncated 537 chars]

Remediation
View remediation

Remediation Suggestions

  1. Replace daxiapi-cli@latest with a specifically reviewed version.
  2. Commit a lockfile containing resolved versions and integrity metadata.
  3. Verify the npm publisher, package provenance, and corresponding source repository before adoption.
  4. Document the exact reviewed source commit and package checksum.
  5. Test dependency upgrades before changing the pinned version.
  6. Run the CLI in a sandbox with minimal filesystem and environment access.
  7. Restrict outbound network access to the API endpoints strictly required for the task.
  8. Avoid running the package as an administrator or other privileged account.
  9. Ensure sensitive environment variables unrelated to the task are not available to the process.

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:42
Finding

Plaintext API Token Exposure Through Verification Output

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:42, 220; references/cli-commands.md:107; references/token-setup.md:38-39
Vulnerability Type: Plaintext sensitive-data disclosure
Risk Level: Medium

bash
npx daxiapi-cli@latest config get token

Relevant documentation excerpt translated into English:

markdown
If the configured Token is returned, the configuration was successful.

Technical Analysis

The documented verification process retrieves and prints the configured API token rather than returning a boolean configuration status or a masked identifier. In an Agent environment, command output can become part of the model context, conversation transcript, tool logs, telemetry, debugging records, or execution history.

This violates the principle that secrets should never be displayed unless disclosure is strictly necessary. Verifying whether a credential exists does not require revealing its complete value.

The exposure is local and output-based in the audited instructions; the project does not contain confirmed code that directly exfiltrates the token to an attacker. Nevertheless, printing the secret creates multiple secondary disclosure paths.

Attack Path

  1. A token has already been stored through the CLI configuration command.
  2. The Agent follows Step 0 or the troubleshooting instructions and executes config get token.
  3. The CLI prints the complete token to standard output.
  4. The tool output is captured in the Agent transcript, execution logs, telemetry, terminal history, or debugging interface.
  5. Any user, operator, extension, plugin, or logging system with access to those records can recover the credential.
  6. The exposed token can then be used against the DaxiAPI service within the permissions and quota assigned to the token.

Impact Assessment

Successful exploitation exposes the user's DaxiAPI credential. An attacker may obtain the API access, subscript ...[truncated 267 chars]

Remediation
View remediation

Remediation Suggestions

  1. Replace config get token with a command that returns only configured or not configured.
  2. If an identifier is needed, display only a short masked fingerprint, such as the final four characters.
  3. Ensure the CLI redacts tokens from standard output, standard error, debug logs, and exception messages.
  4. Prevent secret-bearing tool output from being added to Agent-visible transcripts or telemetry.
  5. Store tokens through an operating-system credential manager or dedicated secret-management service rather than an ordinary plaintext configuration file.
  6. Review and sanitize existing logs that may contain tokens.
  7. Rotate any token that has already been displayed in a conversation, shared terminal, or retained log.
  8. Add automated tests that fail if complete credential values appear in CLI output.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (27)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The trigger phrases are broad finance terms such as '热门股票' and '强势股', which may cause the skill to activate for generic stock-related requests beyond the user's intended scope. Over-broad invocation increases the chance that users are steered into executing the skill's command workflow, including risky CLI and token-setup steps, without sufficient intent or context.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The skill instructs use of npx daxiapi-cli@latest, which fetches and executes the newest package version at runtime. This creates a supply-chain risk: a compromised publisher account, malicious release, or breaking update could execute arbitrary code in the user's environment when the skill is followed.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill instructs users to set an API token but does not include precautions for secret handling, storage, rotation, or avoiding exposure in logs/shell history. In context, this is more dangerous because the token is entered into an unpinned npx command, combining credential-handling risk with supply-chain execution risk.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

This command again relies on @latest, causing runtime download and execution of whatever package version is current. Because this specific step handles token configuration, a malicious or trojanized package could steal credentials or alter local config while appearing to perform normal setup.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
96% confidence
Finding

The skill uses an unpinned npx invocation for market-data retrieval, exposing users to arbitrary code execution if the package supply chain is compromised. Even though the business purpose is benign, the execution model is risky because the command is run directly from the network at the time of use.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
96% confidence
Finding

This repeated @latest CLI usage has the same supply-chain and arbitrary-code-execution exposure as the other invocations. Since users may run multiple such commands, the attack surface and opportunity for compromise increase across the workflow.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

This duplicate occurrence on the same line repeats the insecure @latest pattern in a token-reconfiguration context. If exploited, it could expose or tamper with authentication material while the user is attempting recovery from an error.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The entire template and its required output structure are specified only in Chinese, including mandatory sections and phrasing such as '必须遵守', which effectively constrains the skill to a single language. The file does not offer a user language choice or explain a justified region-specific limitation, which matches the language/locale policy violation criteria.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documentation instructs users to set and retrieve an API token without clearly warning that the token is sensitive credential material. In practice, showing config get token in docs can encourage exposing secrets in terminals, shell history, screen recordings, logs, or shared environments, increasing the chance of credential leakage and unauthorized API use.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The document title and all instructional content are presented only in Chinese, with no indication that the user may choose another language or that the locale restriction is required for a region-specific compliance reason. Under the policy rules, forcing a specific language without opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.