A fast, minimal AI coding agent in Rust. Any OpenAI-compatible provider, persistent sessions, skill-based subagents, and sandboxed shell execution.

Disclaimer: This project is actively developed. While it supports any OpenAI-compatible API, not all models have been thoroughly tested. mlx-community/gemma-4-e4b-it-4bit (Gemma 4) is used as the primary model for development and testing.

# Interactive mode — just start talking
pie
# resume your last conversation in this directory
pie -r # or pie --resume
# Pipe a question
echo "what does src/main.rs do?" | pie --md
# Use a specific model
pie -m mlx-community/gemma-4-e4b-it-4bit

• Persistent sessions — conversations saved per directory, resume with pie -r
• Any provider — works with OpenAI, Anthropic, Groq, Ollama, or any OpenAI-compatible API
• Skills & subagents — markdown-based skills from agentskills.io, auto-loaded from queries
• Native Sandboxing — shell commands run with built-in OS isolation (sandbox-exec on macOS, bubblewrap on Linux)
• Streaming TUI — real-time tool calls, markdown rendering, command history
• Scriptable — --json and --md flags for single-shot mode

# Interactive (default)
pie
# Resume last session
pie -r
# Single-shot output
pie --md "explain this function"
pie --json "list files" # pipe into jq, etc.
# Use a specific skill
pie "/explore summarize this repo"

Pie is configured via environment variables, a pie.toml file, or CLI flags.

The fastest way to get started is with environment variables:

export OPENAI_API_KEY="sk-..."
export OPENAI_MODEL="mlx-community/gemma-4-e4b-it-4bit"
export OPENAI_BASE_URL="http://localhost:1234/v1"
# For providers that support Anthropic-compatible endpoints (e.g. zai)
export ANTHROPIC_BASE_URL="https://api.z.ai/api/anthropic"

For managing multiple providers or project-specific settings, use pie.toml. Pie searches for this file in:

1. ~/.pie/pie.toml (Global configuration)
2. ./.pie/pie.toml (Project-specific configuration)

For a full list of configuration options, see .pie/pie.toml.example.

default_provider = "local"
[provider.local]
model = "mlx-community/gemma-4-e4b-it-4bit"
base_url = "http://localhost:1234/v1"
api_key = "sk-..."
[provider.zai]
model = "glm-5.1"
base_url = "https://api.z.ai/api/paas/v4/"
anthropic_url = "https://api.z.ai/api/anthropic"
api_key = "..."

[agent] max_steps = 25 # Max tool-call iterations per query

To use a specific provider from your config:
```bash
pie -p ollama "how are you?"

• Sandbox: Configure restrictions in pie.toml under [sandbox].
• Skills: Add custom skills to .pie/skills/<name>/SKILL.md.
• Instructions: Add project-level instructions to AGENTS.md.

brew tap dineshdb/pie https://github.com/dineshdb/pie
brew install dineshdb/pie/pie

cargo build --release

MIT