Evot

Building software just got 10× faster and token-efficient.

A self-evolving agent engine — fully observable, built for long-running complex work.

The engine behind evot.ai

News · Why · Install · Quickstart · Dev · Community

export-1776943794877.mp4

📢 News

• 2026-05-08 🛡️ [REPL] /harden — stress-test plans and git changes before shipping. Inspired by @cjzafir.
• 2026-05-02 🧩 [Skills] Builtin skill support — review ships built-in, no install needed.
• 2026-04-28 🖼️ [Image] Resize, preserve through compaction, persist to disk.
• 2026-04-23 🔍 [Search] Full-text session search — /resume <query> to find any past conversation.
• 2026-04-18 📜 [REPL] /history + /goto — time-travel through conversation context.

---

⚡ Why Evot

Most agents dump everything into context — bloated outputs, stale history, invisible decisions. Tokens burn. Quality drifts.

Evot does the opposite:

• Zero-waste context. Every prompt is minimal, high-signal, rebuilt from scratch each turn.
• Half the tokens, half the time. Less noise → fewer turns → complex tasks done faster.
• Self-evolving. Full observability into every LLM call and tool execution feeds back into the engine — each prompt gets leaner automatically.
• Everything searchable. Full-text index over all sessions — /resume <query> to find any past conversation, decision, or code snippet instantly.

Installation

One-liner (recommended)

curl -fsSL https://evot.ai/install | sh

From source

git clone https://github.com/evotai/evot.git
cd evot
make setup && make install
evot

Quickstart

1. Set your API key

Create ~/.evotai/evot.env:

# Anthropic (default)
EVOT_LLM_ANTHROPIC_API_KEY=sk-ant-...
EVOT_LLM_ANTHROPIC_BASE_URL=your-anthropic-base-url
EVOT_LLM_ANTHROPIC_MODEL=claude-opus-4-6
# Multiple models: EVOT_LLM_ANTHROPIC_MODEL=claude-sonnet-4-6,claude-opus-4-6

# Or OpenAI
# EVOT_LLM_OPENAI_API_KEY=sk-...
# EVOT_LLM_OPENAI_BASE_URL=your-openai-base-url/v1
# EVOT_LLM_OPENAI_MODEL=gpt-5.5

# Or DeepSeek (Anthropic-compatible)
# EVOT_LLM_DEEPSEEK_API_KEY=sk-...
# EVOT_LLM_DEEPSEEK_BASE_URL=https://api.deepseek.com/anthropic
# EVOT_LLM_DEEPSEEK_PROTOCOL=anthropic
# EVOT_LLM_DEEPSEEK_MODEL=deepseek-v4-pro

# Or Xiaomi MiMo-V2.5-Pro (Anthropic-compatible)
# EVOT_LLM_XIAOMI_API_KEY=tp-...
# EVOT_LLM_XIAOMI_BASE_URL=https://token-plan-cn.xiaomimimo.com/anthropic
# EVOT_LLM_XIAOMI_PROTOCOL=anthropic
# EVOT_LLM_XIAOMI_MODEL=mimo-v2.5-pro

Use --model provider:model for one-off overrides.

2. Run

evot                                          # interactive REPL
evot -p "summarize today's PRs"               # one-shot task
evot -p "review this" -f ./src/main.rs        # attach file context
evot -p "continue work" -c                   # continue latest session in cwd
evot -p "continue work" -r my-session         # resume or create session

CLI flags & options

Flag	Description
-p, --prompt	Run a single prompt and exit
-f, --file <path>	Attach file/directory context
-c, --continue	Continue the latest session in the current directory
-r, --resume <id>	Resume or create a session
--model <model>	Override the configured model
--verbose	Enable info-level logging

Development

make setup        # install Rust toolchain, git hooks
make test         # all tests (engine + CLI)
make install      # compile standalone binary to ~/.evotai/bin/evot

Community

• GitHub Issues — Bug reports / Feature
• Twitter @Evot_AI — Announcements
• team@evot.ai — Reach the team directly

License

Apache-2.0

---

Built with 🦀 + TypeScript by Evot AI