CowAgentvsMyrlin Workbook
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
CowAgent
zhayujie · Agent harness
OSSMCP
Panel
6.44 spec wins
- Reliability
- 6.0
- Usefulness
- 6.3
- Cost
- 6.5
- Longevity
- 6.7
“The README states that the project does not issue any cryptocurrency, which tells you what three years of its inbox looked like.”
Myrlin Workbook
therealarthur · Agent harness
OSS
Panel
6.42 spec wins
- Reliability
- 5.7
- Usefulness
- 6.3
- Cost
- 8.3
- Longevity
- 5.2
“There is a demo mode, so you can watch somebody else's coding sessions before admitting you want to watch your own.”
Spec by spec
| Spec | CowAgent | Myrlin Workbook |
|---|---|---|
| Architecture | ||
| Category | Agent harness | Agent harness |
| Runs | local | local |
| Platforms | macos, linux, windows | macos, linux, windows, web |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | Yes | No |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | No | YesMyrlin calls no model itself; edits come from the Claude Code or Codex session it spawns in a terminal pane. |
| Git operations | No | YesBranch, dirty state and ahead/behind per project, worktree create and delete, and promoting an issue to a worktree plus a session. |
| Browser control | Yes | No |
| Sandboxed execution | No | No |
| Multi-agent orchestration | Yes | YesSeveral sessions across both CLIs run at once in their own worktrees; they are not coordinated with each other. |
| Headless / CI mode | No | No |
| Models | ||
| Backbone | Claude, OpenAI, Gemini, DeepSeek, Qwen, GLM, Kimi, MiniMax, Doubao, ERNIE, LinkAI | Claude Code, Codex |
| Bring your own model | Yes | Yes |
| Local models | Yes | No |
| Cost | ||
| Pricing model | byok | byok |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | MIT | AGPL-3.0 |
| GitHub stars | 47,207 | 387 |
Which one would each critic pick
| Critic | CowAgent | Myrlin Workbook | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.0 | 6.8 | CowAgent — Pick CowAgent if you want one assistant answering in WeChat, Feishu and DingTalk from a single install; pick QwenPaw if you want a sandbox under the shell tool. |
| El Crítico | 6.0 | 6.0 | no preference |
| El Profesor | 5.8 | 6.5 | Myrlin Workbook — The cost figures are parsed out of transcripts rather than retrieved from a billing interface, which makes them a derived estimate rather than a report. |
| La Inversora | 6.5 | 5.8 | CowAgent — A rename from chatgpt-on-wechat carries 46k stars into a new brand, and the business is LinkAI, which sells hosting, role-based access and private deployment on top. |
| La Jefa | 5.5 | 6.0 | Myrlin Workbook — A board card owning a branch, a worktree and a live session is close to how my teams already split work, and there is no console behind any of it. |
| El Hacker | 7.5 | 7.3 | CowAgent — MIT, a Custom provider row for local models or a proxy, one mcp.json with stdio and SSE that hot-reloads, and skills installable from GitHub or ClawHub with a slash command. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.