Decision Workspace
kan vs jishuken vs causari
Side-by-side comparison of Rust crates
48
kan
experimentalv0.1.0
Local reasoning, global coherence — memory for AI agents.
44
jishuken
experimentalv0.1.0
Jishuken: self-verifying memory for agents, backed by fact files and an op log.
43
causari
experimentalv0.4.1
AI-written code has no author; it has causes. Measures how much AI-tagged code survives in any git repo and records the prompt, model and files behind every agent edit.
Core Metrics
| kan | jishuken | causari | |
|---|---|---|---|
| Health Score | 48 | 44 | 43 |
| Total Downloads | 482 | 13 | 70 |
| 30d Downloads | 0 | 0 | 0 |
| Dependents | 0 | 0 | 0 |
| Releases | 22 | 1 | 5 |
| Last Updated | 48d ago | 28d ago | 1d ago |
| Age | 2m | 28d | 19d |
Health Breakdown
kan
Maintenance
11
Quality
15
Community
6
Popularity
3
Documentation
13
jishuken
Maintenance
11
Quality
11
Community
6
Popularity
1
Documentation
15
causari
Maintenance
14
Quality
9
Community
6
Popularity
2
Documentation
12
Technical Details
| kan | jishuken | causari | |
|---|---|---|---|
| Version | 0.1.0 | 0.1.0 | 0.4.1 |
| Stable (≥1.0) | ✗ No | ✗ No | ✗ No |
| License | MIT | MIT | Apache-2.0 |
| Dependencies | 0 | 13 | 21 |
| Crate Size | 1KB | 101KB | 512KB |
| Features | 0 | 0 | 0 |
| Yanked % | 0.0% | 0.0% | 0.0% |
| Edition | 2024 | 2021 | 2024 |
| MSRV | — | 1.90 | 1.85 |
| Owners | 1 | 1 | 1 |
Links
Quick Verdict
- •kan leads with a health score of 48/100, but none of the options score above 80.