AgentTakt runs on your own machine — the client starts it, so there is no endpoint to ping. 227 installs a week from pypi. Last commit 29 Aug 2026.
Review, edit, and approve AI agent task plans in a ComfyUI-style visual node editor in your terminal
Today is the operative word: we check AgentTakt every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add agenttakt -- uvx agenttakt
{
"mcpServers": {
"agenttakt": {
"args": [
"agenttakt"
],
"command": "uvx"
}
}
}
[mcp_servers.agenttakt]
command = "uvx"
args = ["agenttakt"]
{
"mcpServers": {
"agenttakt": {
"args": [
"agenttakt"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"agenttakt": {
"args": [
"agenttakt"
],
"command": "uvx"
}
}
}
Pin visual feedback on any live web page and send it to Claude Code, Cursor or your AI agent.
Deterministic visual marketing engine. Your agent plans, renders, and posts on-brand campaigns.
See where your cloud and AI bills go, and spend less. Local-first, in your terminal or editor.
Abnormal Security email threats, cases, and reporting in your terminal and your AI agents.
Edit uploaded footage with AI tools: cuts, captions, reframing and previews. Final export in Studio.
Review your AITWIRE recommendations and Brand Canon; file corrections for your approval.
Human-in-the-loop review and approval for AI agents. Audit trail, approval policies, native MCP.
Read and edit octopus.do visual sitemaps: pages, blocks and review notes, as the signed-in user.
Answers built from our own checks of this server.