Skillget

cli

mcp
Guvenlik Denetimi
Basarisiz
Health Uyari
  • License — License: MIT
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 5 GitHub stars
Code Basarisiz
  • rm -rf — Recursive force deletion command in install.sh
Permissions Gecti
  • Permissions — No dangerous permissions requested
Purpose
This CLI and MCP server analyzes your codebase to generate call graphs, dependency maps, and blast radius analyses. It feeds this structural context to AI coding agents like Claude Code and Codex to help them understand your project.

Security Assessment
Overall risk: Medium. The tool requires network requests to function, sending your codebase data to an external API (`api.supermodeltools.com`) via `supermodel analyze` and authenticating via GitHub OAuth. The code is written in Go, which is generally safe, and it does not request dangerous local permissions. However, the automated scan flagged a recursive force deletion command (`rm -rf`) inside the `install.sh` script, which always warrants caution. Users should carefully inspect the installation script before executing it.

Quality Assessment
The project is very new and currently has low community visibility, evidenced by only 5 GitHub stars. However, it is actively maintained (last pushed 0 days ago), properly licensed under the permissive MIT license, and has a clear, detailed README. A notable concern is that several installation and build instructions are currently marked as placeholders, indicating the project is still in early or heavy development.

Verdict
Use with caution — while the project is actively developed and MIT-licensed, its low community adoption, early-stage placeholder instructions, and reliance on uploading your code to an external server require careful review before use.
SUMMARY

CLI for giving AI coding agents a map of your codebase — call graphs, dependency graphs, dead code detection, and blast radius analysis.

README.md

Supermodel CLI

Platform
Website
API

Give your AI coding agent a map of your codebase.

Supermodel CLI connects Claude Code, Codex, and other AI agents to the Supermodel API — providing call graphs, dependency graphs, dead code detection, and blast radius analysis as live context during your sessions.

Links
Website supermodeltools.com
API Docs api.supermodeltools.com
Dashboard dashboard.supermodeltools.com
Twitter / X @supermodeltools
Contact [email protected]

What It Does

Feature Description
Graph pregeneration Analyze your repo upfront so your agent has instant access to call and dependency graphs without waiting mid-task
Dead code detection Surface functions and files with no callers across TypeScript, JavaScript, Python, Go, Rust, and more
Blast radius Before making a change, show which files and functions would be affected downstream
Token efficiency Ship only the graph slices relevant to the current task — not the whole repo — keeping context lean
Agent integration Plug directly into Claude Code, Codex, and Hermes as a context tool

Installation

macOS

# placeholder: brew install supermodeltools/tap/supermodel

Linux

# placeholder: install script
curl -fsSL https://supermodeltools.com/install.sh | sh

From Source

git clone https://github.com/supermodeltools/cli
cd cli
# placeholder: build instructions

Quickstart

1. Authenticate

supermodel login
# Opens your browser for GitHub OAuth

2. Pre-generate your graph

cd /path/to/your/repo
supermodel analyze
# Uploads repo, runs analysis, caches graph locally

3. Use with your agent

Claude Code:

# placeholder: MCP server config or plugin setup
supermodel claude-code install

Codex:

# placeholder: Codex tool integration
supermodel codex install

Hermes:

# placeholder: Hermes integration
supermodel hermes install

Key Commands

supermodel docs                 # Generate static architecture documentation
supermodel analyze              # Analyze the current repo and cache results
supermodel dead-code            # List functions and files with no callers
supermodel blast-radius <file>  # Show what's affected if this file changes
supermodel graph                # Print or export the graph for the current repo
supermodel status               # Show cached graph state and last analysis time
supermodel login                # Authenticate with your Supermodel account
supermodel logout               # Clear stored credentials

How It Works

  1. supermodel analyze zips your repository and uploads it to the Supermodel API.
  2. The API runs static analysis — building a base IR, call graph, and domain classification.
  3. Results are cached locally (and on the API) keyed by a content hash of your repo.
  4. Your agent tool integration reads from the cache and injects the relevant graph slice into context.

Graph data is never sent to your AI provider directly — only the slices your agent requests.

Supported Agents

Agent Status
Claude Code Planned — #1
Hermes Planned — #2
Codex Planned — #3

Pricing

Usage is metered per analysis. Run supermodel status to check your balance.

Contributing

Issues and PRs welcome. See CONTRIBUTING.md for guidelines.

Questions? Open an issue or email [email protected].

Yorumlar (0)

Sonuc bulunamadi