headlong vs picobot
Persistent-agency agent harness in ~10K lines of Bash: it keeps thinking between messages, thinks by writing shell commands, and one shared mind serves a whole team over Slack or Telegram. — versus — A self-hosted personal AI agent in a single ~9MB Go binary — persistent memory, 16 tools + MCP, skills, cron/heartbeat, and Telegram/Discord/Slack/WhatsApp channels. Runs on a $5 VPS.
Both are self-hosted always-on personal agents with persistent memory and IM channels. Picobot is a 9MB Go binary with 16 fixed tools that runs on a $5 VPS; Headlong is Bash all the way down, has no tool system, and thinks continuously instead of on triggers.
| headlong | picobot | |
|---|---|---|
| Stars | 1.0k | 1.3k |
| Forks | 95 | 166 |
| Language | Shell | Go |
| License | Apache-2.0 | MIT |
| Last activity | today | 5 months ago |
| Topics | agents, memory | agents |
| Curated connections | 6 | 4 |
headlong — the curator's take
Read this if you think harnesses have converged. Headlong drops the tool system entirely — the agent thinks by writing Bash, so `curl` is the HTTP client and `jq` the JSON parser — and it never stops: your message lands in an ongoing thought stream as one more observation, and the agent decides whether to answer. The trajectory is a fork/merge DAG of jsonl files, and context is a projection of it at exponentially decaying resolution, so nothing is compacted away in place. Laude Institute research alpha: it runs real shell commands around the clock at $1-2/hour, so use a spend-capped key and let the installer put it in Docker. Not the pick if you want a reactive, per-user, request/response coding agent — that mode exists but the whole design is aimed elsewhere.
picobot — the curator's take
The anti-bloat statement piece: zero dependencies, ~10MB RAM idle, instant cold start — an always-on agent with ranked memory recall, background subagents, a natural-language HEARTBEAT.md cron, and self-authored skills ('create a skill for checking weather' → it writes the markdown), reachable from your phone via four chat channels. Runs on a Raspberry Pi or Termux on an old Android. Any OpenAI-compatible endpoint, including Ollama. Its own README names the target: OpenClaw's power without the 500MB container. NOT a coding harness and single-user by design — this is your personal daemon, not a team platform; complex multi-agent workflows will outgrow it fast, which is rather the point.