Codex CLIvsPoolside
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
Codex CLI
OpenAI · Terminal agent
#1OSSMCP
Panel
7.11 spec wins
- Reliability
- 7.3
- Usefulness
- 7.0
- Cost
- 6.8
- Longevity
- 7.2
“Apache-2.0, so you can read every line except the one that lets you pick a model that is not OpenAI's.”
Poolside
poolside · Terminal agent
#15MCP
Panel
6.90 spec wins
- Reliability
- 6.8
- Usefulness
- 7.0
- Cost
- 6.5
- Longevity
- 7.2
“One of its three network policies is named unsafe-allow-all, which is at least the most clearly labelled decision on this board.”
Spec by spec
| Spec | Codex CLI | Poolside |
|---|---|---|
| Architecture | ||
| Category | Terminal agent | Terminal agent |
| Runs | local | local, sandbox, cloud |
| Platforms | macos, linux, windows | macos, linux, windows |
| Context window | not documented | 1M tokens on Laguna S 2.1, 256k on XS 2.1 and M.1 |
| Protocols | ||
| MCP client | Yes | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | Yes | Yes |
| Browser control | NoThe shared Codex codebase carries a browser-use configuration with Chrome DevTools Protocol access, but OpenAI's documentation states browser is not available in the Codex CLI or the IDE extension; it runs in ChatGPT on the web and in the desktop app. The CLI's --search flag adds live web search only. | NoWeb search and fetch are documented; there is no DOM-level browser automation. |
| Sandboxed execution | Yes | YesTool commands run inside a container on your machine and need a local Docker engine; network policy is off, allow-list or unsafe-allow-all, enforced through a proxy container. |
| Multi-agent orchestration | Yes | Yes |
| Headless / CI mode | Yes | Yes`pool exec --prompt ... --output json` with POOLSIDE_API_KEY, plus a documented GitHub Actions integration. |
| Models | ||
| Backbone | OpenAI GPT-5.x, GPT-5.x-Codex | Laguna S 2.1, Laguna XS 2.1, Laguna M.1 |
| Bring your own model | Yes | Yes |
| Local models | Yes | YesDocumented local-run guides for Ollama and vLLM on Metal, plus full on-premises and air-gapped deployment. |
| Cost | ||
| Pricing model | mixed | mixed |
| Starts at | $8/mo | n/a |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | No |
| License | Apache-2.0 | proprietary |
| GitHub stars | 127,563 | n/a |
Which one would each critic pick
| Critic | Codex CLI | Poolside | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 8.0 | 6.8 | Codex CLI — If you already pay for ChatGPT, this is the terminal agent you have and it is a good one; open source, MCP, headless, and pointed at OpenAI until you configure it otherwise. |
| El Crítico | 7.0 | 6.5 | Codex CLI — A sandbox mode you choose once is the only barrier between the agent and your shell, the free tier is a taste, and the defaults point at one vendor. |
| El Profesor | 6.3 | 6.5 | Poolside — The vendor trains its own model family and publishes no evaluation of it anywhere on the row, which is a conspicuous silence for a company selling its own inference. |
| La Inversora | 8.3 | 7.0 | Codex CLI — A lab's terminal agent bundled into a $20 subscription with 121,000 stars on the repo; the distribution is ChatGPT, and that is the whole thesis. |
| La Jefa | 7.3 | 7.0 | Codex CLI — Business at $20 a seat with SAML SSO and no default training, Enterprise for SCIM and audit logs; if the org is on ChatGPT already, this is a line item, not a project. |
| El Hacker | 5.8 | 7.5 | Poolside — The client is proprietary and the model weights are published under OpenMDW-1.1 and Apache-2.0, which is the exact opposite of everyone else on this board. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.