herdrvsNezha
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
herdr
herdr · Agent harness
OSS
Panel
6.31 spec wins
- Reliability
- 6.2
- Usefulness
- 5.7
- Cost
- 7.5
- Longevity
- 6.0
“A tmux for agents whose detach key is ctrl+b q, so twenty years of muscle memory finally counts as onboarding.”
Nezha
hanshuaikang · Agent harness
OSS
Panel
6.33 spec wins
- Reliability
- 5.7
- Usefulness
- 6.0
- Cost
- 8.2
- Longevity
- 5.3
“It writes your commit messages too, so eventually nothing in the repository will be in your own words.”
Spec by spec
| Spec | herdr | Nezha |
|---|---|---|
| Architecture | ||
| Category | Agent harness | Agent harness |
| Runs | local | local |
| Platforms | macos, linux, windows | macos, linux, windows |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | No | No |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | No | Yes |
| Git operations | No | Yes |
| Browser control | No | No |
| Sandboxed execution | No | No |
| Multi-agent orchestration | Yes | YesSeveral Claude Code and Codex sessions run at once across projects; Nezha manages them rather than coordinating them into one task. |
| Headless / CI mode | No | No |
| Models | ||
| Backbone | Claude Code, Codex, Cursor, OpenCode, Grok | via managed agents (Claude Code, Codex) |
| Bring your own model | No | NoModel choice belongs to the underlying Claude Code or Codex installation, which must be installed before Nezha is used. |
| Local models | No | No |
| Cost | ||
| Pricing model | byok | byok |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | No | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | Apache-2.0 | GPL-3.0 |
| GitHub stars | 41,859 | 1,900 |
Which one would each critic pick
| Critic | herdr | Nezha | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.0 | 6.8 | herdr — Pick herdr if your Claude Code, Codex, Cursor or OpenCode sessions die when you close the laptop; pick Claude Squad if what you actually need is a worktree per task. |
| El Crítico | 6.0 | 6.0 | no preference |
| El Profesor | 6.8 | 6.3 | herdr — One Rust binary with a socket API, per-pane status tracking, and a wait primitive that returns only when another agent is genuinely blocked; no benchmark, and nothing to benchmark. |
| La Inversora | 4.8 | 6.0 | Nezha — An individual author with 1,892 stars and no company behind the project, which is interest without a business, and the roadmap depends on two vendors he does not own. |
| La Jefa | 5.5 | 5.8 | Nezha — Sixty desktop installations, each one needing the agent CLI installed first, and nothing that runs unattended, so it never becomes a step I can measure. |
| El Hacker | 8.0 | 7.0 | herdr — Apache-2.0, one binary via brew or mise, a socket API I can script from anything, no MCP and no model settings because it owns terminals, not agents. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.