@vera-ai/cli 0.12.10 → 0.12.12

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/README.md +22 -29
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -12,40 +12,33 @@ Current benchmark snapshot: on Vera's local 21-task, 4-repo release benchmark, `
12
12
  npm install -g @vera-ai/cli
13
13
  ```
14
14
 
15
- `vera setup` with no flags runs an interactive wizard (backend + agent skills + optional indexing). For focused changes: `vera backend` manages the ONNX runtime, `vera agent install` manages skill files and can update `AGENTS.md` / `CLAUDE.md` style project instructions, and `vera agent sync` refreshes stale skills after upgrades.
16
-
17
- ## Usage
15
+ ## Quick Start
18
16
 
19
17
  ```bash
20
- # Optional: install skill files for your agents
21
- vera agent install
22
-
23
- # Index a project
18
+ vera setup --api
24
19
  vera index .
25
-
26
- # Search
27
- vera search "authentication middleware"
28
-
29
- # Local ONNX inference (no API keys needed. downloads models automatically)
30
- vera index . --onnx-jina-cpu
31
- vera search "error handling" --onnx-jina-cpu
32
-
33
- # Optional local CodeRankEmbed preset
34
- vera setup --code-rank-embed --onnx-jina-cuda
35
-
36
- # GPU acceleration (NVIDIA/AMD/DirectML/CoreML/OpenVINO)
37
- vera index . --onnx-jina-cuda
38
-
39
- # Diagnose or repair local setup issues
40
- vera doctor
41
- vera doctor --probe
42
- vera repair
43
- vera upgrade
20
+ vera search "authentication logic"
44
21
  ```
45
22
 
46
- `vera doctor --probe` runs a deeper read-only ONNX session check. `vera upgrade` shows the binary update plan and can apply it when the install method is known.
47
-
48
- On GPU backends, Vera uses a free-VRAM-aware batch ceiling and sequence-aware local micro-batching, and it reuses learned device-specific batch windows across runs.
23
+ `vera setup` with no flags runs an interactive wizard. `vera agent install` manages skill files for your coding agents and can update `AGENTS.md` / `CLAUDE.md` style project instructions.
24
+
25
+ ## Common Tasks
26
+
27
+ | Task | Command |
28
+ |------|---------|
29
+ | Use the interactive setup wizard | `vera setup` |
30
+ | Use API mode (recommended) | `vera setup --api` |
31
+ | Use a local NVIDIA backend | `vera setup --onnx-jina-cuda` |
32
+ | Search semantically | `vera search "authentication middleware"` |
33
+ | Keep the index up to date | `vera update .` |
34
+ | Watch for file changes | `vera watch .` |
35
+ | Diagnose setup issues | `vera doctor` |
36
+ | Run the deeper ONNX probe | `vera doctor --probe` |
37
+ | Repair missing local assets | `vera repair` |
38
+ | Inspect binary upgrades | `vera upgrade` |
39
+ | Install agent skills | `vera agent install` |
40
+
41
+ For the full backend matrix, model options, Docker setup, and troubleshooting, see the main [README](https://github.com/lemon07r/Vera) and [Installation Guide](https://github.com/lemon07r/Vera/blob/master/docs/installation.md).
49
42
 
50
43
  ## What you get
51
44
 
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@vera-ai/cli",
3
- "version": "0.12.10",
3
+ "version": "0.12.12",
4
4
  "description": "Bootstrap installer and wrapper for the Vera CLI",
5
5
  "license": "MIT",
6
6
  "homepage": "https://github.com/lemon07r/Vera#readme",