Decision Workspace
claude-agent-rs vs capo-agent vs agent-harness
Side-by-side comparison of Rust crates
51
claude-agent-rs
experimentalv1.1.0
Lightweight AI coding agent in Rust — runs Claude on your Max subscription for $0. 8 built-in tools, MCP client, hooks, skills with frontmatter, sessions, sub-agents, streaming.
48
capo-agent
experimentalv0.11.2
Coding-agent library built on motosan-agent-loop. Composable, embeddable.
47
agent-harness
experimentalv0.3.5
Drive LLM coding agents — Claude Code, OpenAI Codex, and local Ollama / OpenAI-compatible models — from Rust behind one trait, with a normalized streaming event vocabulary. Bring your own agent too.
Core Metrics
| claude-agent-rs | capo-agent | agent-harness | |
|---|---|---|---|
| Health Score | 51 | 48 | 47 |
| Total Downloads | 203 | 687 | 202 |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 54 | 0 |
| Releases | 12 | 27 | 12 |
| Last Updated | 103d ago | 8d ago | 6d ago |
| Age | 4m | 1m | 1m |
Health Breakdown
claude-agent-rs
Maintenance
14
Quality
15
Community
6
Popularity
3
Documentation
13
capo-agent
Maintenance
12
Quality
10
Community
11
Popularity
3
Documentation
12
agent-harness
Maintenance
14
Quality
12
Community
6
Popularity
3
Documentation
12
Technical Details
| claude-agent-rs | capo-agent | agent-harness | |
|---|---|---|---|
| Version | 1.1.0 | 0.11.2 | 0.3.5 |
| Stable (≥1.0) | ✓ Yes | ✗ No | ✗ No |
| License | MIT | MIT | MIT OR Apache-2.0 |
| Dependencies | 15 | 28 | 5 |
| Crate Size | 67KB | 132KB | 74KB |
| Features | 0 | 1 | 4 |
| Yanked % | 0.0% | 0.0% | 8.3% |
| Edition | 2021 | 2021 | 2021 |
| MSRV | — | 1.80 | 1.74 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •claude-agent-rs leads with a health score of 51/100, but none of the options score above 80.
- •capo-agent has the most downloads (687), suggesting wider adoption.
- •capo-agent, agent-harness are pre-1.0 — API may change.