Skip to content
Agentmart
Submit a tool

AgentTakt

Review, edit, and approve AI agent task plans in a ComfyUI-style visual node editor in your terminal

Connector by ryoohshima · GitHub account verified by the MCP Registry

IndexedMITLocal (stdio)★ 1
Write a review

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/ryoohshima-agenttakt

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Review, edit, and approve AI agent task plans in a ComfyUI-style visual node editor in your terminal

Reviews

Write a review

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools
1 read · 1 write · 0 delete
Secrets
None asked for
Install scripts
Not scanned yet
Runs as
Python package

Tools by what they do 2

Read from the source code; the running server may list different tools.

Read 1 · Look things up without changing anything.

  • show_plan

Write 1 · Create, change, send or run things.

  • request_approval

Tools (2)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

  • request_approval

    Submit a task execution plan for human review in the AgentTakt editor. Blocks until the human approves or rejects the plan in the TUI. Returns {"status": "approved" | "rejected", "plan": <edited plan>, "reason": <str | null>}. The returned plan may differ from the submitted one (the human can edit nodes, edges and parameters) — always execute the returned plan. Args: plan: Plan JSON with graph_id, nodes[] and edges[] (must be a DAG). summary: One-line description shown in the editor header.

  • show_plan

    Display a task execution plan in the AgentTakt editor, without blocking. Call this whenever you have formulated a multi-step plan — in any mode, not just plan mode — so the human can see it as a node graph. Returns {"status": "displayed"} as soon as the editor receives the plan; it does NOT wait for (or report) any human decision. Use request_approval instead when you need the human's approval before executing. Args: plan: Plan JSON with graph_id, nodes[] and edges[] (must be a DAG). summary: One-line description shown in the editor header.

Signals

Stars
1
Forks
0
Open issues
9
Weekly downloads
66
Last commit
Sep 23, 2026 (1 month ago)
Last release
Sep 23, 2026 (1 month ago)
Last package publish
Sep 21, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name
io.github.ryoohshima/agenttakt
Version
0.2.2
Auth
No auth
Packages
pypi agenttakt
License
MIT
Repository
github.com/ryoohshima/AgentTakt
Website
github.com/ryoohshima/AgentTakt
First seen
Oct 24, 2026
Last checked
Oct 24, 2026
Source
Official MCP Registrygithub-graphql
Cite this page

AgentTakt MCP server. Agent Mart. https://agentmart.dev/connectors/ryoohshima-agenttakt (last updated 2026-10-24)