askimo.chat · AI for your workflows, with the freedom to choose any provider.

Askimo

Askimo is a provider-agnostic AI toolkit that brings powerful AI capabilities to your command line and automation workflows.
From chatting with LLMs to automating DevOps pipelines, Askimo connects to any provider - OpenAI, X AI, Gemini, Anthropic, or local models like Ollama.

AI for your workflows, with the freedom to choose any provider.

---

✨ Why Askimo

• Provider Freedom
  Switch between OpenAI, Gemini, X AI, Anthropic, or Ollama with the same commands. No vendor lock-in as the AI landscape evolves.

• Automation-First Design
  Built for DevOps and automation workflows. Pipe files, logs, or command output into Askimo and let AI handle analysis, transformation, and decision-making.

• RAG-Enabled Projects
  Create intelligent project workspaces with built-in vector search (pgvector). Your AI assistant knows your codebase, documentation, and project context.

• Reusable Recipes
  Build and share parameterized AI workflows. Create templates for code reviews, log analysis, documentation generation, and more.

• Dual Interface
  Choose your workflow: interactive chat for exploration, or non-interactive mode perfect for scripts, CI/CD pipelines, and automation.

• Extensible Platform
  Add custom providers, commands, and integrations. Askimo grows with your team's needs.

---

🎬 Demo

Summarizing files, generating commit messages, and integrating with Git

Chatting with multiple AI providers

---

🧠 Core Capabilities

AI Chat

• Interactive conversations with multiple AI providers

Knowledge Management

• RAG-enabled projects with automatic document indexing
• Vector search powered by PostgreSQL + pgvector
• Project workspaces that give AI context about your codebase
• Contextual responses based on your project's files and documentation

Automation & DevOps

• Pipeline-friendly non-interactive mode for CI/CD integration
• Recipe system for reusable, parameterized AI workflows
• Log analysis and system monitoring with AI insights
• Stdin/stdout support for seamless integration with existing tools

Platform Features

• Provider-agnostic architecture (OpenAI, Gemini, X AI, Ollama)
• Extensible plugin system for custom providers and commands
• Configuration management with per-provider parameter tuning

---

⚙️ Quickstart

macOS / Linux (Homebrew)

brew tap haiphucnguyen/askimo
brew install askimo
askimo

Windows (Scoop)

scoop bucket add askimo https://github.com/haiphucnguyen/scoop-askimo
scoop install askimo
askimo

Other ways to install → Installation Guide

👉 Once installed, you can connect Askimo to providers like Ollama, OpenAI, Gemini, or X AI and start chatting.

📖 See Getting Started for tutorials on setting up Ollama, adding API keys (OpenAI, Gemini, X AI), switching providers, and running real workflow examples.

---

💬 CLI Usage

Askimo supports two modes for running commands:

Interactive Mode

Start Askimo without arguments to enter interactive mode:

askimo
askimo> :help

Non-Interactive Mode

Run commands directly from the command line:

askimo --help
askimo --list-providers
askimo --set-provider openai
echo "function add(a, b) { return a + b; }" | askimo -p "Convert this to TypeScript"

Direct Chat (Non-Interactive)

Send a single message to AI without entering interactive mode:

askimo -p "Your prompt here"
askimo --prompt "Your prompt here"

With piped input:

# Analyze code from stdin
echo "function add(a, b) { return a + b; }" | askimo -p "Convert this to TypeScript"

# Process file contents
cat myfile.js | askimo --prompt "Explain this code"

# Analyze git changes
git diff | askimo -p "Summarize these changes"

# Process command output
ls -la | askimo -p "Explain these file permissions"

---

🧩 Extending Askimo

Askimo is designed to be pluggable, so you can tailor it to your needs:

• Add a new chat model → Creating a New Chat Model
• Create a new command → Creating a New Command

---

🤝 Contributing

• Fork & clone the repo
• Create a feature branch
• Open a PR