OpenClawvsPaperclip
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
OpenClaw
OpenClaw Foundation · Agent harness
OSSMCP
Panel
6.89 spec wins
- Reliability
- 6.0
- Usefulness
- 6.8
- Cost
- 7.3
- Longevity
- 7.2
“Warelay, Clawdbot, Moltbot, OpenClaw: three renames in ten weeks, one of them courtesy of Anthropic's trademark lawyers.”
Paperclip
Paperclip Labs, Inc. · Agent harness
OSS
Panel
6.70 spec wins
- Reliability
- 6.2
- Usefulness
- 6.7
- Cost
- 7.0
- Longevity
- 7.0
“Gives every agent a title and a reporting line, so the first thing your AI workforce receives is middle management.”
Spec by spec
| Spec | OpenClaw | Paperclip |
|---|---|---|
| Architecture | ||
| Category | Agent harness | Agent harness |
| Runs | local, sandbox | local, cloud |
| Platforms | macos, linux, windows | macos, linux, web |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | Yes | No |
| MCP server | Yes | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | No |
| Git operations | No | No |
| Browser control | Yes | No |
| Sandboxed execution | Yes | No |
| Multi-agent orchestration | Yes | Yes |
| Headless / CI mode | No | No |
| Models | ||
| Backbone | OpenAI, Anthropic, OpenRouter, GitHub Copilot, Ollama, LM Studio, vLLM, llama.cpp | via agent adapters (Claude Code, Codex, Gemini, Cursor, OpenClaw, Hermes, Pi, HTTP and shell agents) |
| Bring your own model | Yes | No |
| Local models | Yes | No |
| Cost | ||
| Pricing model | byok | free |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | No |
| Openness | ||
| Open source | Yes | Yes |
| License | MIT | MIT |
| GitHub stars | 391,182 | 95,916 |
Which one would each critic pick
| Critic | OpenClaw | Paperclip | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.3 | 7.0 | OpenClaw — Pick OpenClaw if you want an always-on assistant on your own hardware that answers on Telegram; pick Hermes Agent if you want it to teach itself new skills. |
| El Crítico | 6.5 | 6.0 | OpenClaw — Sandboxing is off by default, the docs call it not a perfect security boundary even when on, and bind mounts walk straight through it. |
| El Profesor | 7.0 | 6.5 | OpenClaw — A gateway on the host routes tool calls into per-agent containers over Docker, Podman, SSH or OpenShell, and MCP runs in both directions; no benchmark is published. |
| La Inversora | 6.3 | 6.3 | no preference |
| La Jefa | 5.3 | 6.3 | Paperclip — Monthly budgets per agent with hard stops and spend tracked by company, project and provider is the first orchestrator here with a finance answer; SSO and support are still missing. |
| El Hacker | 8.8 | 8.3 | OpenClaw — MIT, an mcp.servers block with per-server toolFilter, Ollama, LM Studio, vLLM and llama.cpp as first-class providers, and it serves MCP too; this one is mine. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.