Back to skill

Security audit

Seedream AI Image Generator

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Media.io image-generation connector, but it sends prompts, image URLs, and a Media.io API key to Media.io when used.

Install this only if you are comfortable sending your generation prompts, reference image URLs, and Media.io API key to Media.io and potentially spending account credits. Keep the API key out of logs and shared prompts, prefer an isolated environment, and pin dependencies if you package this for repeated use.

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

T08 · Insecure Dependencies

Warning
Location
SKILL.md:64
Finding

Unpinned Third-Party Dependency Installation

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 64–67
Vulnerability Type: Unpinned dependency installation
Risk Level: Medium

Vulnerable Code Snippet:

markdown
### 1) Install Dependency
```bash
pip install requests
text

### Technical Analysis

The installation instructions retrieve `requests` without specifying an exact version, a lock file, or cryptographic hashes. Consequently, the package version and its transitive dependency graph are mutable and determined by the package index configured on the user's system at installation time.

This creates a software supply-chain risk. A compromised package index, malicious internal mirror, dependency-confusion condition, or compromised future release could cause attacker-controlled code to be installed. Python packages can execute code during installation in applicable build workflows and when imported during normal Skill execution.

### Attack Path

1. An attacker compromises or controls a package source accessible through the user's pip configuration, or compromises a package release in the unresolved dependency chain.
2. The user follows the documented command `pip install requests`.
3. Pip resolves and installs the mutable package version and dependencies without verifying an approved lock file or integrity hashes.
4. Attacker-controlled code executes during a package build or installation, or when `requests` is imported by `scripts/skill_router.py`.
5. The malicious code operates with the privileges of the user or service account running the installation or Skill.

### Impact Assessment

Successful exploitation could permit arbitrary code execution under the privileges of the user or service account running pip or the Skill. Depending on those privileges and the execution environment, an attacker could access application data and environment variables such as `API_KEY`, alter local files, make unauthorized network requests, or compromise o
...[truncated 271 chars]
Remediation
View remediation

Remediation Suggestions

  • Replace the unconstrained installation command with a reviewed, exactly pinned dependency set.
  • Maintain direct and transitive dependencies in a lock file.
  • Record and enforce cryptographic hashes with a command such as:
    bash
    pip install --require-hashes -r requirements.txt
    
  • Generate the requirements file from a trusted environment and review dependency changes before updating it.
  • Configure pip to use the official Python Package Index or an authenticated, trusted internal mirror.
  • Install dependencies in an isolated virtual environment or container using a non-privileged account.
  • Add automated dependency vulnerability and provenance checks to the release process.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill invokes environment access for API_KEY, reads local files such as scripts/c_api_doc_detail.json, and performs outbound network requests, but the manifest does not declare an explicit tool/permission scope. This weakens reviewability and least-privilege controls because users and hosting systems cannot easily see or enforce what capabilities the skill needs before execution.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill sends user prompts, optional image URLs, and an API credential to a third-party service, but the documentation does not clearly warn that user-supplied content will leave the local environment and be processed by Media.io. This creates privacy and data-handling risk, especially if users provide sensitive prompts or private image URLs without informed consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The documentation instructs users to send an API key to an external service but does not include any warning about credential sensitivity, third-party transmission, or safe handling. In an agent-skill context, this increases the risk that users or downstream tooling will expose secrets or unknowingly authorize outbound requests to a remote provider.

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/c_api_doc_detail.json (reported line 12)May include surrounding context.

json
"description": "API to query user credits balance.",
    "api_header": "{\"list\": [{\"name\": \"X-API-KEY\", \"value\": \"API key to authorize requests\"}, {\"name\": \"Content-Type\", \"value\": \"application/json\"}], \"title\": \"Authorizations\", \"describe\": \"Add the following authorization information in the request header\"}",
    "api_body": "{\"title\": \"Request Body\", \"category\": [{\"list\": [], \"title\": \"Query Credits\", \"describe\": \"Request body to query user credits balance\"}]}",
    "api_request_demo": "{\"title\": \"Example Request\", \"request\": [{\"title\": \"Query User Credits\", \"language\": \"cURL\", \"code_example\": \"curl --request POST \\\\n  --url https://openapi.media.io/user/credits \\\\n  --header 'Content-Type: application/json' \\\\n  --header 'X-API-KEY: <api-key>' \\\\n  --data '{}'\"}]}",
    "api_response": "{\"list\": [{\"name\": \"code\", \"type\": \"integer\", \"describe\": \"Response status code, 0 indicates success\"}, {\"name\": \"msg\", \"type\": \"string\", \"describe\": \"Response message, empty string on success\"}, {\"name\": \"data\", \"type\": \"object\", \"describe\": \"Response data object\"}, {\"name\": \"credits\", \"type\": \"integer\", \"describe\": \"User credits balance, located within the data object\"}], \"title\": \"Response\", \"describe\": \"After the request is successfully processed, the server will return the following response\"}",
    "api_code_demo": "{\"list\": [{\"code\": \"0\", \"describe\": \"Success\"}, {\"code\": \"40001\", \"describe\": \"Invalid API key\"}, {\"code\": \"40002\", \"describe\": \"API key expired\"}], \"title\": \"Status Code\"}",
    "content": null,

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The image-generation API documentation states that image URLs and prompts are sent to an external Media.io endpoint, but it does not warn users that potentially sensitive images or text prompts will leave the local environment. For an agent skill, that omission can cause unintentional disclosure of private data, copyrighted material, or confidential business content to a third party.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

This code file performs a network request to an external service and transmits the contents of body, which is derived from params, but there is no confirmation prompt, print/log message, or warning comment/docstring explaining that user or system data may be sent off-host. The method docstring describes invocation generically and does not disclose the outbound data transmission behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.