AgentTasker MCP Server

io.github.S3bRRPublished by GitHub account S3bRRv1.0.1Updated Oct 24, 2026

Minimal stdio MCP server for parallel task execution by AI agents.

VerifiedSTDIODesktop onlyWeb Search & ScrapingDeveloper Tools

Overview

AI-generated overview

A local stdio MCP server that lets an AI assistant run web searches, fetch webpages, and call other MCP tools in parallel batches.

What you can do
Run single or batched web searches and webpage fetches, with Markdown or local HTML extraction and optional hosted rendering fallback for difficult sites. Execute batches of calls to other MCP tools configured through a separate server config file, with concurrency chosen by the agent up to a set limit. Start background batches, poll them for results, and cancel queued work. List the remote tools available from the configured MCP servers. Cache search results and apply rate limiting with compact results.
When to use it
Useful when an assistant needs to gather many pages or searches at once, or fan out repeated calls to other MCP tools, without doing them one at a time. Also suited to longer jobs that should run in the background while the assistant continues.
Requirements
Local process on the user's machine; desktop only. Python 3.10+ and uv/uvx, or a Git clone with the setup script. First launch needs network access. Web search requires a search provider: Brave with the BRAVE_SEARCH_API_KEY environment variable, or a running SearXNG instance with JSON enabled. Calling other MCP tools requires a separate servers config file passed as an argument. No API key is needed for plain webpage fetching.
Before you install
Only connect MCP servers you trust, since AgentTasker opens its own stdio connections to them and remote tool calls are not cached or retried because they may have side effects. The optional Jina reader fallback shares the fetched URLs with Jina and is off by default. Cancelling a batch skips queued work but already-running calls may finish. Background jobs and caches are process-local and disappear on restart. Provider quotas and rate limits still apply.

Installation

In SourceWeft

  1. Open AgentTasker MCP Server in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

This repository has no README.

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v1.0.1LatestSep 16, 2026

Score

25% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts

    Declare a GitHub repository with a README, or serve resource_documentation in the OAuth protected-resource metadata.

  • License10 pts

    Declare a license in the repository or the registry entry.