Back to skill

Security audit

yintai-task-runner

Security checks for vulnerabilities and agentic risk

Overview

This skill can automatically claim paid tasks and mark them completed, but the inspected implementation only creates placeholder deliverables rather than performing the requested work.

Review before installing. Use only low-privilege Yintai credentials, do not pass secrets on the command line, keep TASK_API_BASE_URL fixed to a trusted HTTPS endpoint, avoid cron or continuous mode until execution is real and validated, and do not let it claim paid or production tasks unless placeholder completion is acceptable.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (5)

T09 · Insecure Skill Coding Practices

Error
Location
executor.py:52
Finding

Placeholder Deliverables Are Falsely Reported as Completed Tasks

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
executor.py:58
Finding

Deliverable Upload Failures Are Suppressed and Reported as Successful Execution

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
api_client.py:87
Finding

API Credentials Can Be Exfiltrated Through an Unrestricted Configurable Base URL

Content
View full analysis
str: """构建完整URL""" return f"{self.base_url}{self.config.api_prefix}{path}" def _get_headers(self) -> dict: """ 生成认证请求头 Returns: 请求头字典 """ headers = { "Content-Type": "application/json", } # API Key 和 Secret 认证 if self.api_key: headers["X-API-Key"] = self.api_key if self.api_secret: headers["X-API-Secret"] = self.api_secret ``` ```python # api_client.py:112-139 async def _request( self, method: str, path: str, params: dict = None, json_data: dict = None, ) -> dict: """发送认证请求""" url = self._make_url(path) headers = self._get_headers() async with httpx.AsyncClient() as client: if method.upper() == "GET": response = await client.get(url, params=params, headers=headers) elif method.upper() == "POST": response = await client.post(url, json=json_data, headers=headers) elif method.upper() == "PUT": r ...[truncated 2105 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
skill.py:221
Finding

API Secrets Are Accepted Through Exposed Command-Line Arguments

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
install.sh:61
Finding

Installer Uses Unpinned Dependencies and an Unquoted Shell Version Constraint

Content
View full analysis
=0.25.0 ``` ```toml # pyproject.toml:27-29 dependencies = [ "httpx>=0.25.0", ] ``` ### Technical Analysis The installation process upgrades `pip` to an unspecified latest version and installs any `httpx` version greater than or equal to `0.25.0`. There is no lock file, exact version pin, or package-hash verification. Installation results can therefore change without any modification to the audited Skill. In the shell script, `httpx>=0.25.0` is not quoted. A POSIX-compatible shell interprets `>` as an output-redirection operator. The effective command may install unconstrained `httpx` and redirect output to a file named `=0.25.0`, rather than passing the intended version constraint as one package argument. Although no malicious dependency was identified in the audited files, this pattern unnecessarily increases supply-chain exposure and makes reviewed builds non-reproducible. ### Attack Path 1. A user runs `install.sh`, as required by the Skill metadata and documentation. 2. The script downloads the current package installer and current matching dependency versions from the configured package index. 3. The unquoted redirection character prevents the intended lower-bound expression from being reliably passed to `pip`. 4. A compromised package-index account, malicious mirror, dependency compromise, or newly vulnerable release is selected during installation. 5. Package installation executes dependency build or installation logic inside the created environment. 6. The compromised package runs whenever the Skill imports it. This is a supply-chain risk path; the audit did not confirm that ...[truncated 604 chars]
Remediation
View remediation
" ``` 2. Use the virtual environment's interpreter explicitly instead of relying on the ambient `pip` executable. 3. Pin all direct and transitive dependencies in a reviewed lock file. 4. Verify packages with cryptographic hashes, such as by using `pip --require-hashes`. 5. Avoid unconditional installer upgrades; pin a reviewed `pip` version when an upgrade is required. 6. Use a trusted, authenticated package index and prohibit unexpected extra indexes. 7. Generate a software bill of materials and scan locked dependencies for known vulnerabilities. 8. Correct the stray-file behavior caused by shell redirection and test the installer in a clean environment. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (30)

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The documented behavior claims broad task-grabbing and task-management capabilities, but the detected implementation behavior appears centered on generating and uploading deliverables, with key control flows missing. This mismatch is dangerous because users and reviewers may authorize the skill based on one trust model while it actually performs a different, less transparent set of actions involving external APIs and file creation.

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 271)May include surrounding context.

md
- CLI 详细用法、Cron 配置:见 `./references/usage.md`

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The README explicitly promotes autonomous polling, task grabbing, status updates, and deliverable upload to a remote service, but does not clearly warn users that enabling the skill will cause ongoing remote actions and outbound data transmission. In an agent-skill context, that omission is security-relevant because users may invoke or install the skill without understanding that it can continuously act on external systems and automatically upload artifacts.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
96% confidence
Finding

The skill declares no explicit tool scope or allowed-tools despite documented use of environment variables, filesystem access, network requests, and shell execution. This creates overbroad operational authority and makes it harder to constrain or audit what the skill may access at runtime, especially since it handles credentials and uploads artifacts.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The documentation instructs users to obtain API credentials from a local OpenClaw config file, expanding the skill's effective behavior into local secret discovery. Directing an agent or user to read secrets from unrelated local configuration introduces credential-exposure risk and broadens access beyond the core task-running purpose.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill documents use of sensitive credentials from a local config source without any warning about safe secret handling, masking the sensitivity of the operation. In practice this increases the chance that secrets are read, echoed, logged, or mishandled during troubleshooting or normal use.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

Natural-language instructions telling the agent to retrieve API credentials from a local config file and use them for subsequent operations create a clear path for secret access and misuse. In an agent context, such instructions can effectively turn documentation into a prompt-level mechanism for harvesting local secrets and using them in network actions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The module docstring describes the skill entirely in Chinese, with no indication that other languages are supported or that this locale restriction is intentional and justified. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The client adds X-API-Key and X-API-Secret headers to outbound HTTP requests, which is a credential-handling and network-transmission operation. Although the code has an internal comment about authentication, there is no user-facing warning, confirmation, or visible disclosure that sensitive credentials will be sent to the remote API.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This method reads a local ZIP file and uploads both the file contents and result_description to a remote endpoint. That is a data-transmission operation affecting user data, but the code provides no user-facing warning, confirmation prompt, or explicit disclosure that local files and report text will be sent off-host.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The module docstrings and user-facing default message are written entirely in Chinese, including the default delivery message at L27. This indicates a language-specific behavior without any visible opt-in, fallback, or justification that the skill is intended only for Chinese-speaking users or a China-specific environment.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The code sends the generated ZIP and result description through an API client, which likely transmits task content and metadata over the network. While upload failures are logged, there is no user-facing warning or confirmation that local task data will be uploaded to a remote system.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

This code writes task metadata and task content, including identifiers, title, category, creator ID, and description, into local files. There is no confirmation prompt or user-facing disclosure in this file warning that task data will be persisted to disk before packaging.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

This is a manifest file, so vague-trigger rules apply. The description says the skill will 'automatically poll the task system, grab orders, and execute' but does not specify when it should activate, under what commands, or any exclusions, making invocation scope ambiguous.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The API documentation states that uploading a deliverable automatically changes the task status to completed, but this side effect is only noted briefly and without an explicit warning or separate confirmation requirement. In a task-runner skill that automates grabbing, executing, and delivering tasks, this increases the risk of accidental or premature completion, which can irreversibly alter workflow state and trigger downstream business effects.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The documentation explicitly demonstrates supplying API credentials via command-line flags, which commonly exposes secrets through shell history, process listings, terminal logging, and CI/job logs. In this skill's context, those credentials appear to grant access to a task system, so accidental exposure could allow unauthorized task access or manipulation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documentation presents all operational instructions and configuration descriptions in Chinese, which effectively forces a specific language for users. The file does not indicate that the skill is region-specific or provide any user opt-in or alternative language guidance.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The natural-language content of the skill documentation is entirely Chinese, and there is no indication that users may choose another language or that the locale restriction is intentional and justified. Under SQP-3, forcing a specific language without user opt-in is a policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

Most of the operational instructions and descriptions are written only in Chinese, with no indication that users may choose another language or locale. Under SQP-3, forcing a specific language without user opt-in can constitute a natural-language policy violation when no justification or alternative is provided.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The file’s user-facing docstring is written entirely in Chinese ('CLI 入口', '支持') with no indication that other languages are supported. This can conflict with language/locale policy when a skill presents a fixed language without user opt-in or documented justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The file-level description and user-visible/docstring text are written in Chinese, indicating an implicit language choice with no sign of user opt-in or locale selection. The policy requires avoiding forced language or locale constraints unless documented and justified or made optional.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The title and user-facing messages in this file are written in Chinese, and the generated completion message also assumes Chinese output. This can violate language/locale policy when the skill forces a specific language without user opt-in or documented justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

Multiple user-facing messages, including errors and completion instructions, are presented only in Chinese. This creates a language-policy issue because the skill forces a specific language on users without opt-in or justification for a region-specific audience.

Content

No source excerpt is available for this finding.

External Script Fetching

Low
Category
Supply Chain
Confidence
15% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · install.sh (reported line 37)May include surrounding context.

sh
if ! python3 -m pip --version &> /dev/null; then
        echo "❌ pip 未安装,尝试安装..."
        python3 -m ensurepip --upgrade 2>/dev/null || {
            echo "请手动安装 pip: curl https://bootstrap.pypa.io/get-pip.py | python3"
            exit 1
        }
    fi

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
73% confidence
Finding

The natural-language description is presented only in Chinese, with no indication that users can choose another language or that the skill is intentionally limited to a Chinese-speaking context. This can violate language/locale policy when no opt-in or justification is provided.

Content

No source excerpt is available for this finding.