mirror of
https://github.com/farcasclaudiu/openclaw.git
synced 2026-06-29 13:02:10 +03:00
chore: Also format scripts and skills.
This commit is contained in:
@@ -6,21 +6,21 @@ Core programs that ship with OpenProse. Production-quality, well-tested programs
|
||||
|
||||
### Evaluation & Improvement
|
||||
|
||||
| Program | Description |
|
||||
|---------|-------------|
|
||||
| `inspector.prose` | Post-run analysis for runtime fidelity and task effectiveness |
|
||||
| `vm-improver.prose` | Analyzes inspections and proposes PRs to improve the VM |
|
||||
| Program | Description |
|
||||
| ------------------------ | -------------------------------------------------------------- |
|
||||
| `inspector.prose` | Post-run analysis for runtime fidelity and task effectiveness |
|
||||
| `vm-improver.prose` | Analyzes inspections and proposes PRs to improve the VM |
|
||||
| `program-improver.prose` | Analyzes inspections and proposes PRs to improve .prose source |
|
||||
| `cost-analyzer.prose` | Token usage and cost pattern analysis |
|
||||
| `calibrator.prose` | Validates light evaluations against deep evaluations |
|
||||
| `error-forensics.prose` | Root cause analysis for failed runs |
|
||||
| `cost-analyzer.prose` | Token usage and cost pattern analysis |
|
||||
| `calibrator.prose` | Validates light evaluations against deep evaluations |
|
||||
| `error-forensics.prose` | Root cause analysis for failed runs |
|
||||
|
||||
### Memory
|
||||
|
||||
| Program | Description |
|
||||
|---------|-------------|
|
||||
| `user-memory.prose` | Cross-project persistent personal memory |
|
||||
| `project-memory.prose` | Project-scoped institutional memory |
|
||||
| Program | Description |
|
||||
| ---------------------- | ---------------------------------------- |
|
||||
| `user-memory.prose` | Cross-project persistent personal memory |
|
||||
| `project-memory.prose` | Project-scoped institutional memory |
|
||||
|
||||
## The Improvement Loop
|
||||
|
||||
@@ -40,6 +40,7 @@ The evaluation programs form a recursive improvement cycle:
|
||||
```
|
||||
|
||||
Supporting analysis:
|
||||
|
||||
- **cost-analyzer** — Where does the money go? Optimization opportunities.
|
||||
- **calibrator** — Are cheap evaluations reliable proxies for expensive ones?
|
||||
- **error-forensics** — Why did a run fail? Root cause analysis.
|
||||
@@ -84,11 +85,13 @@ prose run lib/project-memory.prose --backend sqlite+
|
||||
The memory programs use persistent agents to accumulate knowledge:
|
||||
|
||||
**user-memory** (`persist: user`)
|
||||
|
||||
- Learns your preferences, decisions, patterns across all projects
|
||||
- Remembers mistakes and lessons learned
|
||||
- Answers questions from accumulated knowledge
|
||||
|
||||
**project-memory** (`persist: project`)
|
||||
|
||||
- Understands this project's architecture and decisions
|
||||
- Tracks why things are the way they are
|
||||
- Answers questions with project-specific context
|
||||
|
||||
Reference in New Issue
Block a user