Atlas / MCP servers / dnnyngyen / Iron Manus

Iron ManusBLOCK

Name: Iron Manus
Author: dnnyngyen

mcp/dnnyngyen/iron-manus-1

MCP server by dnnyngyen · claude · claude code · manus · Audited 2026-10-07

⚑ Report / re-scan Source ↗

Iron Manus MCP

Verdict

BLOCK

Grade

Trust score

69 /100

Exposed tools

22 22r · 0w · 0d

Transport

stdio

License

MIT

Stars

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

[](https://mseep.ai/app/dnnyngyen-iron-manus-mcp)

[](https://github.com/dnnynguyen/iron-manus-mcp/actions) [](https://github.com/dnnynguyen/iron-manus-mcp/actions) [](https://www.typescriptlang.org/) [](https://nodejs.org/) [](https://opensource.org/licenses/MIT) [](https://hub.docker.com/r/dnnyngyen/iron-manus-mcp) [](https://github.com/dnnynguyen/iron-manus-mcp/pkgs/container/iron-manus-mcp) [](https://github.com/dnnynguyen/iron-manus-mcp)

Model Context Protocol server for AI workflow orchestration inspired by Manus AI's orchestration patterns and Andrej Karpathy's "Iron Man" analogy.

Extended context management through agent delegation and Todos as a custom agent dispatch tool

Demo

🎥 Video Tutorial

[](https://www.youtube.com/watch?v=EFVQT3pmyTc)

[Watch the detailed walkthrough](https://www.youtube.com/watch?v=EFVQT3pmyTc)

Historical Notes: Architectural Patterns

Note: This project is now archived. The following notes document architectural decisions that later appeared in mainstream AI tooling.

Several patterns implemen

Read from source at commit 73d7a2888460OBSERVED · 2026-10-07

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code

claude mcp add iron-manus-mcp -- npx -y [email protected]

claude-desktop

{
  "mcpServers": {
    "iron-manus-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "[email protected]"
      ]
    }
  }
}

Exposed tools (22)

22 read · 0 write · 0 destructive.

Tool	Risk	Description
`AmiiboAPI`	read	Nintendo Amiibo information and database
`CountAPI`	read	Free counting service for tracking page hits and events
`GeoJS`	read	IP geolocation with country and city information
`IP-API`	read	IP geolocation and ISP information
`JokeAPI`	read	Programming, miscellaneous, and dark humor jokes
`Pokemon-3D-api`	read	3D Pokemon models for ThreeJS - GLB format with regular and shiny variants
`analyzer`	read	Data analysis, metrics analysis, performance analysis, insights
`architecture.svg`	read	Architecture diagram for ${topic}
`coder`	read	Implementation, programming, development, building applications
`critic`	read	Quality assessment, security review, code review, validation
`custom_styles.css`	read	Custom styles for ${topic} presentation
`ipapi.co`	read	IP geolocation API with detailed location data
`logo.png`	read	Company or project logo
`planner`	read	Strategic planning, architecture design, system planning, project management
`process_flow.svg`	read	Process flow diagram for ${topic}
`researcher`	read	Information gathering, knowledge synthesis, documentation research
`slide_generator`	read	Unified slide generation with template-based specialization (12 templates available)
`synthesizer`	read	Integration, optimization, combining systems, workflow coordination
`title_background.jpg`	read	Background image for ${topic} presentation title
`ui_architect`	read	UI architecture, design systems, component architecture, interface design
`ui_implementer`	read	UI implementation, component building, frontend development
`ui_refiner`	read	UI refinement, styling, aesthetics, polish, optimization

Trust audit

BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

Layer	What it checks	Result
L0	Provenance & inventory	PASS
L1	Static analysis of the code	FAIL
L2	Instruction surface (what it tells the agent)	PASS
L3	Class-specific surface	PASS
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (6 observation(s))
Network: declared (3 observation(s))
Shell: declared (7 observation(s))
Dependencies: not all pinned
Secrets in source: none-found

Findings (11)

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

src/utils/python-security.ts:58

/eval\s*\(/i, // eval() calls

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95

src/utils/python-security.ts:59

/exec\s*\(/i, // exec() calls

Why it matters. evaluates text as code

Fix. remove; use a parser or a dispatch table