Aria2 Agent
@Rehui-2006
About Aria2 Agent
AI Agent-first download tool built on aria2. Forces file-allocation=none, uses RPC tellStatus for reliable completion detection. MCP compatible, zero dependencies.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"aria2-agent": {
"command": "python",
"args": [
"-m",
"aria2cli",
"mcp"
]
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is Aria2 Agent?
Aria2 Agent is a CLI tool and MCP (Model Context Protocol) server that wraps aria2’s JSON-RPC interface with safety rules for AI Agents. It solves the false-positive completion problem caused by aria2’s default preallocation, giving agents a reliable download status via pure JSON output. It is built for developers and AI agent users who need trustworthy, large-file downloads.
How to use Aria2 Agent?
Install with pip install aria2-agent or download the single-file aria2cli.py. Ensure Python 3.8+ and aria2c are in PATH. Start the RPC daemon once per session (python aria2cli.py start --dir ~/Downloads), then submit downloads with python aria2cli.py add <URL> --dir <path> --out <file> --wait --timeout 3600. The agent reads the JSON result’s complete field. For MCP integration, run python aria2cli.py mcp and configure the server in Claude Desktop or Cursor/VS Code.
Key features of Aria2 Agent
- Forces
file-allocation=noneto prevent phantom files. - Completion judged via RPC only, never local file size.
- Pure JSON output for easy agent parsing.
- Built-in 3x retry, resume, and 3600s timeout.
- MCP Server mode with 8 tools, zero configuration.
- Zero dependencies – pure Python 3.8 standard library.
- Cross-platform: Windows, Linux, macOS.
- Prevents rule circumvention with forbidden per-task options.
Use cases of Aria2 Agent
- Downloading large AI models and datasets with reliable completion reporting.
- Automating downloads in AI agent workflows (Claude, GPT, Cursor).
- Multi-source mirror downloads with automatic fallback.
- Async download management – submit, poll, pause, resume tasks.
FAQ from Aria2 Agent
How is Aria2 Agent different from using aria2c directly?
Aria2 Agent solves the false-positive completion problem where aria2c’s default preallocation creates a full-size placeholder file before data arrives. Agents checking file size get a false “complete” signal. Aria2 Agent forces file-allocation=none, checks completion via RPC only, and outputs clean JSON with a reliable complete field. It also enforces 10 architecturally‑enforced iron rules that cannot be bypassed.
What are the dependencies and runtime requirements?
Python 3.8 or higher and aria2c in PATH. The base tool has zero external dependencies – it uses only the Python standard library. For MCP Server mode, install with pip install "aria2-agent[mcp]".
Where does my data and configuration live?
The RPC daemon listens only on 127.0.0.1. The auto-generated RPC secret token is stored in ~/.aria2_agent/state.json. No data ever leaves your machine. All downloaded files are written to the directory you specify.
Does Aria2 Agent collect telemetry or phone home?
No. The only network connection is to the local aria2 daemon on 127.0.0.1. There is no telemetry, analytics, update checks, or data collection. The entire source is a single 959‑line file for easy auditing.
What transport and authentication does it use?
Aria2 Agent communicates with the local aria2 daemon via JSON
Frequently asked questions
How is Aria2 Agent different from using `aria2c` directly?
Aria2 Agent solves the false-positive completion problem where `aria2c`’s default preallocation creates a full-size placeholder file before data arrives. Agents checking file size get a false “complete” signal. Aria2 Agent forces `file-allocation=none`, checks completion via RPC only, and outputs clean JSON with a reliable `complete` field. It also enforces 10 architecturally‑enforced iron rules that cannot be bypassed.
What are the dependencies and runtime requirements?
Python 3.8 or higher and `aria2c` in PATH. The base tool has zero external dependencies – it uses only the Python standard library. For MCP Server mode, install with `pip install "aria2-agent[mcp]"`.
Where does my data and configuration live?
The RPC daemon listens only on `127.0.0.1`. The auto-generated RPC secret token is stored in `~/.aria2_agent/state.json`. No data ever leaves your machine. All downloaded files are written to the directory you specify.
Does Aria2 Agent collect telemetry or phone home?
No. The only network connection is to the local aria2 daemon on `127.0.0.1`. There is no telemetry, analytics, update checks, or data collection. The entire source is a single 959‑line file for easy auditing.
What transport and authentication does it use?
Aria2 Agent communicates with the local aria2 daemon via JSON
Basic information
More MCP servers
Vibgrate MCP
VibgrateQuery your team's drift, vulnerability, and migration data from any AI assistant. Vibgrate MCP connects Cursor, Claude, ChatGPT, Windsurf, or VS Code to Vibgrate Cloud: 51 tools for DriftScores, CVEs and EOL runtimes, up

Coinrule
Coinrule IncThe Coinrule MCP is an official Model Context Protocol server that connects AI assistants, Claude, ChatGPT, Grok, and others, directly to your Coinrule Cloud account https://cloud.coinrule.com/mcp

LeadMarina
LeadMarinaVerified local-business leads: search any niche in any city, query your lead library, and export to Close, GoHighLevel, Google Sheets, or emailed files — every email SMTP-checked, every phone verified with line type and
AptiBuild AI — Career Intelligence
ParklandBuildsAptiBuild AI MCP Server — career, labor market, and business idea intelligence for AI agents. 11 tools covering real-time FRED labor market signals, BLS wage data (126 occupations), AI-generated app ideas, career paths,

Kilo IoT Platform
Kilo IoTYour AI can finally reach the physical world. Kilo is an AI-first IoT platform for commercial and industrial operations, and this connector hands a live deployment to the AI client you already use. Not an export. Not a
Comments