Back to skill

Security audit

Aria2 Downloader

Security checks for vulnerabilities and agentic risk

Overview

The skill does what it claims, but it handles an aria2 control token in ways that could leak it if users rely on the documented protections.

Review before installing. Use only an HTTPS aria2 RPC endpoint or a trusted local-only endpoint, protect .aria2-config.json yourself with git ignore rules and restrictive file permissions, and treat the token as granting control over your aria2 download queue. Be careful with remove commands because they act on the specified GID without an extra confirmation step.

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 (2)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/aria2.js:41
Finding

Aria2 RPC Credentials Can Be Transmitted over Plaintext HTTP

Content
View full analysis
{ const body = JSON.stringify({ jsonrpc: '2.0', id: Date.now(), method: method, params: ['token:' + config.token, ...params] }); const url = new URL(config.url); const isHttps = url.protocol === 'https:'; const lib = isHttps ? https : http; const options = { hostname: url.hostname, port: url.port || (isHttps ? 443 : 80), path: url.pathname || '/jsonrpc', method: 'POST', headers: { 'Content-Type': 'application/json', 'Content-Length': Buffer.byteLength(body) } }; ``` ### Technical Analysis The RPC authentication token is placed directly into every JSON-RPC request as `token:`. The implementation selects Node.js's plaintext `http` client whenever the configured URL does not use the `https:` protocol. No validation rejects insecure HTTP endpoints or limits them to loopback interfaces. Consequently, a configuration such as `http://aria2.example.com/jsonrpc` transmits the authentication token, RPC method, and method parameters without transport encryption. An attacker with access to the network path can passively capture the token. An active man-in-the-middle attacker could additionally inspect or modify requests and responses. The 15-second request timeout does not provide confidentiality, integrity, or endpoint authentication. ### Attack Path 1. A user configures the Skill with an `http://` aria2 JSON-RPC endpoint. 2. The user invokes `add`, `list`, `pause`, `unpause`, `remove`, or `batch`. 3. The script constructs a request containing `token:`. 4. The script sends the request through Node.js's plaintext HTTP client. 5. An attacker mo ...[truncated 956 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:17
Finding

Secret Configuration Is Claimed to Be Git-Ignored Without a Corresponding Rule

Content
View full analysis
⚠️ 此文件包含敏感信息,已在 `.gitignore` 中排除。 ``` The final sentence states that the file contains sensitive information and has been excluded through `.gitignore`. ### Technical Analysis The Skill instructs users to store the aria2 RPC token in `.aria2-config.json` inside the workspace and explicitly claims that this file is excluded by `.gitignore`. However, the audited project directory contains no `.gitignore` file. Users may reasonably rely on this claim and assume that the credential file cannot be staged. In a Git-managed workspace, the file remains unprotected unless an exclusion rule is supplied elsewhere by the user's environment. Because the configuration contains a reusable bearer-style aria2 RPC secret, accidental repository inclusion can expose operational control of the configured aria2 service. This issue is caused by insecure and misleading secret-management guidance rather than hardcoding a credential in the distributed source. ### Attack Path 1. A user follows the documented setup and creates `.aria2-config.json` in a Git-managed workspace. 2. The user relies on the statement that the file is already excluded by `.gitignore`. 3. Because this project does not provide the claimed ignore rule, the file is visible to Git as an untracked file. 4. The user stages or commits the workspace broadly, for example with `git add .`. 5. The configuration file is pushed to a shared or public repository. 6. A repository reader extracts the RPC URL and token. 7. If the aria2 endpoint is reachable, the reader authenticates to it and exercises the privileges available to that token. ### Impact Assessm ...[truncated 513 chars]
Remediation
View remediation
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

This markdown file documents a remove command for download tasks, but it does not include any warning that the action may be destructive or irreversible for the task state. Under the markdown-specific warning criterion, behaviors affecting user data or system state should be disclosed clearly.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script's comments, errors, help text, and operational messages are consistently presented in Chinese, including all user-visible CLI output. The file does not offer a language/locale option or document that the skill is intentionally limited to Chinese-speaking users.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The removeTask function issues the destructive aria2.remove RPC call and only logs after the deletion has already been requested. There is no confirmation prompt, pre-action warning, or explanatory comment/docstring indicating the destructive effect of removing a task.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The user-facing README content is entirely in Chinese and does not offer any language choice or explain a region-specific constraint. This can violate language or locale policy when a skill imposes a specific language without user opt-in.

Content

No source excerpt is available for this finding.