🎯 Goal
Convert the kdm cache command from static list/get/purge CLI operations into an interactive cache browser where users can browse, preview, and manage cached AI explanations.
📌 Current Behavior
kdm cache manages locally cached AI explanations to save tokens and avoid duplicate network requests.
Subcommands:
| Subcommand |
Description |
kdm cache list |
List all cached explanation keys |
kdm cache get <key> |
Retrieve and print a cached explanation |
kdm cache remove <key> |
Delete a specific cache entry |
kdm cache purge |
Delete all cached explanations |
Example usage:
kdm cache list
kdm cache purge
🚀 Proposed Interactive Dashboard
Layout
Cache Browser 3 entries
──────────────────────────────────────────────────────
> Run-001 openai/gpt-4o 2 hours ago
Run-002 ollama/llama3.1 1 day ago
Run-003 openai/gpt-4o 3 days ago
──────────────────────────────────────────────────────
ENTER:View DELETE:Remove P:Purge All Q:Quit
Preview Pane (press Enter)
┌─ Run-001 ─────────────────────────────────────────┐
│ │
│ Cause: │
│ Container exceeded memory limit of 256Mi. │
│ │
│ Fix: │
│ kubectl edit deployment app │
│ Increase memory: 256Mi -> 512Mi │
│ │
│ [ESC] Close │
└───────────────────────────────────────────────────┘
Purge Confirmation (press P)
┌─────────────────────────────────┐
│ Delete all cached entries? │
│ │
│ [Y] Yes [N] No │
└─────────────────────────────────┘
User Flow
- User runs
kdm cache → list of cached entries loads.
- Use
↑ / ↓ to select a cache entry.
- Press
Enter to preview the cached explanation in an overlay.
- Press
Delete or d to remove the selected entry.
- Press
p to purge all — confirmation dialog appears.
Keyboard Shortcuts
| Key |
Action |
↑ / ↓ |
Navigate cache entries |
Enter |
Preview cached explanation |
Delete / d |
Remove selected entry |
p |
Purge all (with confirmation) |
Esc |
Close overlay |
Q |
Quit |
Recommended Libraries
📋 Contribution Workflow
- Fork the repo and create a feature branch (e.g.,
feature/interactive-kdm-cache).
- Implement the interactive TUI for
kdm cache using the layout and flow described above.
- Submit a PR explicitly linked to this issue (e.g.,
Closes #141).
- 📸 Include a screenshot or recording of the implemented UI in your PR to verify the new dashboard experience.
🎯 Goal
Convert the
kdm cachecommand from static list/get/purge CLI operations into an interactive cache browser where users can browse, preview, and manage cached AI explanations.📌 Current Behavior
kdm cachemanages locally cached AI explanations to save tokens and avoid duplicate network requests.Subcommands:
kdm cache listkdm cache get <key>kdm cache remove <key>kdm cache purgeExample usage:
🚀 Proposed Interactive Dashboard
Layout
Preview Pane (press
Enter)Purge Confirmation (press
P)User Flow
kdm cache→ list of cached entries loads.↑/↓to select a cache entry.Enterto preview the cached explanation in an overlay.Deleteordto remove the selected entry.pto purge all — confirmation dialog appears.Keyboard Shortcuts
↑/↓EnterDelete/dpEscQRecommended Libraries
ink-select-inputchalk📋 Contribution Workflow
feature/interactive-kdm-cache).kdm cacheusing the layout and flow described above.Closes #141).