initrunner
Active·★ 40·Apache-2.0·Updated 2026-10-16
InitRunner lets you define an agent in one YAML file, chat with it, run it autonomously, and deploy it as a daemon triggered by cron, file changes, webhooks, or Telegram messages. It supports multiple execution modes, built-in memory, cost controls, multi-agent orchestration, and security features. Built on PydanticAI.
#agent-framework#ai-agents#ai-automation#ai-tools#autonomous-agents#cli#llm#llm-agents
01
Features
01One file, four modes: interactive REPL, one-shot prompt, autonomous loop, and daemon with triggers.
02Autonomous execution with task decomposition, reasoning strategies, and guardrails (iteration, token, time budgets).
03Daemon mode with six trigger types: cron, webhook, file_watch, heartbeat, Telegram, Discord.
04Built-in memory (semantic, episodic, procedural) that persists across sessions and agents.
05Security features: input validation, tool authorization (InitGuard), sandboxed code execution, tamper-evident audit trail, encrypted credential vault.
02
Compatibility
anthropic
Anthropic
Verified via docs
openai
OpenAI
Verified via docs
google
Google
Verified via docs
groq
Groq
Verified via docs
mistral
Mistral
Verified via docs
cohere
Cohere
Verified via docs
03
Quick start
1
$ curl -fsSL https://initrunner.ai/install.sh | sh
2
$ initrunner setup
04
Use cases
↳Automated code review: set up a daemon that reviews pull requests or file changes.
↳Personal research assistant: create an agent that researches topics, summarizes findings, and stores knowledge.
↳Customer support Q&A: ingest documentation and deploy as a helpdesk bot on Telegram or webhook.
05
Alternatives
nab★ 17
Token-lean web microfetch for LLM agents: any URL → clean markdown via CLI, MCP server, and Claude Code plugin. Real browser-cookie auth, passkeys, anti-bot reach, on-by-default prompt-injection defense, plus on-device multimodal ASR/OCR. A single Rust binary — not a browser.
crow★ 26
Modular, agentic framework and MCP platform you self-host. Build and run your own AI agents, connect Claude/ChatGPT/Cursor as a native MCP connector, self-host your apps, and share over encrypted P2P — on hardware you own, with local or cloud models. Open source.
agentdescent★ 199
Gradient descent, but the parameters are agents — a parallel, asynchronous framework for self-evolving agents (skills, prompts, harnesses). Diffs are the gradients; the aggregator is the optimizer.
jev-judge-mcp★ 49
Typed judgment tools for MCP agents. TypeSafe's Jev model as verify, screen, find, classify, rerank, decide, compare, extract, review, gate, and score: the model judges, policy decides auto, review, or escalate.
ruvos★ 28
rUvOS — the agentic operating system. A single static Rust binary that runs as an MCP server, giving Claude Code persistent memory, resumable signed sessions, multi-agent orchestration, a knowledge graph, safety guardrails, and a signed audit trail. Zero Node.js, zero SQLite, zero daemon.
mcp-servers-live★ 13
Auto-updated index of MCP servers shipping on GitHub, refreshed every 15 minutes
Comments
Log in to leave a comment
No comments yet. Be the first!