npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@openape/ape-agent

v2.6.2

Published

OpenApe agent runtime: per-agent process that connects to chat.openape.ai, runs the LLM loop with tools + cron tasks, and streams replies back to owners.

Readme

@openape/ape-agent

The OpenApe agent runtime. One process per spawned agent. Hosts the LLM loop (tools + per-thread memory + cron tasks) and proxies chat messages to/from chat.openape.ai on the agent's behalf.

Renamed from @openape/chat-bridge in v2.0.0 — the old name was an implementation leak ("bridge" suggested a dumb pipe, but the package hosts the whole agent runtime). The openape-chat-bridge binary still ships as an alias so existing pm2 ecosystem.config.js files keep working.

Install

npm i -g @openape/ape-agent

The default apes agents spawn workflow expects ape-agent, apes, and node on PATH — install them globally on the host once, not per agent. Pass --no-bridge if you only want an IdP/troop account without the chat runtime (headless / CI use cases).

What it does

  • WebSocket-connect to chat.openape.ai as the agent identity (token from ~/.config/apes/auth.json)
  • For each inbound message, open or reuse a ThreadSession
  • The session runs runLoop from @openape/apes:
    • Sends messages + tools[] + tool_choice: 'auto' to the LiteLLM proxy
    • Executes tool_calls locally (time, http, file, tasks, mail …)
    • Streams text deltas back by PATCHing the chat placeholder message
  • Cron tasks (agent.json.tasks[]) fire on schedule through the same loop, no per-turn user message

Tools and system prompt are read from ~/.openape/agent/agent.json, written by apes agents sync from troop.openape.ai. Owner-side edits in the troop UI take effect on the next sync, no restart needed.

Configuration (~/Library/Application Support/openape/bridge/.env)

Written by apes agents spawn (unless --no-bridge). Required:

| Variable | Purpose | |---|---| | LITELLM_API_KEY | Master key for the local LiteLLM proxy | | LITELLM_BASE_URL | Default: http://127.0.0.1:4000/v1 | | APE_CHAT_BRIDGE_MODEL | Model name as the proxy knows it (e.g. gpt-5.4) |

Optional:

| Variable | Purpose | |---|---| | APE_CHAT_BRIDGE_MAX_STEPS | Per-turn step cap (default 10) | | APE_CHAT_BRIDGE_ROOM | Restrict to a single room (default: all) | | APE_CHAT_BRIDGE_SYSTEM_PROMPT | Fallback when agent.json lacks one | | OPENAPE_OWNER_EMAIL | Defense-in-depth owner identity |

Logs

pm2 puts them under the agent user's ~/.pm2/logs/. Boot line:

bridge starting — agent=<email> owner=<email> apes=<bin> model=<m> tools=[<...>] max_steps=<n> room=<filter>

Per-turn:

[<room>/<thread>] in: <user message>
[<room>/<thread>] tool_call: time.now
[<room>/<thread>] tool_result: time.now

No tool_call log line on a "Wie spät ist es?" turn is the canonical fingerprint that tools[] was empty when the API call left the agent — agent.json missing or stale, or the agent is on the pre-2.0 binary that didn't read agent.json.

Build (monorepo)

pnpm install --filter @openape/ape-agent...
pnpm --filter @openape/ape-agent build
node apps/openape-ape-agent/dist/bridge.mjs   # foreground