@vivantel/virage-cli 0.1.51 → 0.1.52

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 +35 -27
  2. package/package.json +1 -1
package/README.md CHANGED
@@ -12,44 +12,50 @@ npx @vivantel/virage-cli <command>
12
12
 
13
13
  ## Commands
14
14
 
15
+ Every top-level command has a single-key alias — use `virage --help` to see all of them.
16
+
15
17
  ### Pipeline
16
18
 
17
- | Command | Description |
18
- |---------|-------------|
19
- | `virage index` | Run the RAG indexing pipeline |
20
- | `virage check` | Validate that the current embedder config matches the stored index |
21
- | `virage validate` | Validate config without running the pipeline |
19
+ | Command | Alias | Description |
20
+ |---------|-------|-------------|
21
+ | `virage index` | `i` | Run the RAG indexing pipeline |
22
+ | `virage check` | `c` | Validate that the current embedder config matches the stored index |
23
+ | `virage validate` | `val` | Validate config without running the pipeline |
24
+ | `virage query <text>` | `q` | Semantic search over the indexed knowledge base |
22
25
 
23
26
  ### Agent integration
24
27
 
25
- | Command | Description |
26
- |---------|-------------|
27
- | `virage init` | Generate a `virage.config.json` template and configure agent plugins |
28
- | `virage update` | Update virage ecosystem packages and re-sync agent plugin config files |
29
- | `virage usage` | Print per-prompt token usage for the current Claude Code session (zero inference tokens) |
30
- | `virage read-skill-summary <name>` | Print the summary for a named Virage skill (used by agent hooks for proactive context injection) |
28
+ | Command | Alias | Description |
29
+ |---------|-------|-------------|
30
+ | `virage init` | — | Generate a `virage.config.json` template and configure agent plugins |
31
+ | `virage update` | `up` | Update all virage packages (reads `virage.config.json`; works in non-Node projects) |
32
+ | `virage install-hooks` | `hooks` | Install git hooks for auto-indexing on pull/branch switch |
33
+ | `virage usage` | `use` | Print per-prompt token usage for the current Claude Code session |
34
+ | `virage read-skill-summary <name>` | `skill` | Print the summary for a named Virage skill |
31
35
 
32
36
  ### Evaluation
33
37
 
34
- | Command | Description |
35
- |---------|-------------|
36
- | `virage evaluate` | Evaluate retrieval quality against an eval dataset |
37
- | `virage eval-generate` | Generate an eval dataset from existing chunks |
38
- | `virage experiment run` | Run an experiment and save results |
39
- | `virage experiment list` | List saved experiment runs |
40
- | `virage experiment compare` | Compare two experiment runs with bootstrap significance test |
41
- | `virage report` | Show observability report from pipeline runs |
38
+ All evaluation commands live under the `virage eval` (`e`) parent:
39
+
40
+ | Command | Alias | Description |
41
+ |---------|-------|-------------|
42
+ | `virage eval run` | `e run` | One-shot retrieval quality check (precision@5/10, MRR, hit rate) |
43
+ | `virage eval generate` | `e gen` | Generate an eval dataset from existing indexed chunks |
44
+ | `virage eval save --name <n>` | — | Run evaluation and save results under a name for later comparison |
45
+ | `virage eval list` | — | List saved evaluation runs |
46
+ | `virage eval compare --baseline --candidate` | — | Bootstrap significance test between two saved runs |
47
+ | `virage report` | `r` | Show observability report from pipeline runs |
42
48
 
43
49
  ### Diagnostics
44
50
 
45
- | Command | Description |
46
- |---------|-------------|
47
- | `virage store stats` | Show vector index quality metrics |
48
- | `virage store perf` | Show query performance report |
49
- | `virage benchmark embedder` | Benchmark embedder latency and throughput |
50
- | `virage chunks report` | Show chunk cohesion report |
51
- | `virage viz embeddings` | Generate a 2D visualization of the embedding space |
52
- | `virage dashboard` | Start a local RAG monitoring dashboard |
51
+ | Command | Alias | Description |
52
+ |---------|-------|-------------|
53
+ | `virage store stats` | — | Show vector index quality metrics |
54
+ | `virage store perf` | — | Show query performance report |
55
+ | `virage benchmark embedder` | — | Benchmark embedder latency and throughput |
56
+ | `virage chunks report` | — | Show chunk cohesion report |
57
+ | `virage viz embeddings` | — | Generate a 2D visualization of the embedding space |
58
+ | `virage dashboard` | `d` | Start a local RAG monitoring dashboard |
53
59
 
54
60
  ### Telemetry
55
61
 
@@ -67,3 +73,5 @@ npx @vivantel/virage-cli <command>
67
73
  All commands accept `-v` / `-vv` / `-vvv` (up to `-vvvvv`) to increase log verbosity.
68
74
 
69
75
  Most pipeline commands accept `-c, --config <path>` to specify a config file (default: `./virage.config.json`).
76
+
77
+ `virage dashboard` additionally accepts `--port <n>` (default: 3000) and `--verbose` (log all HTTP requests).
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@vivantel/virage-cli",
3
- "version": "0.1.51",
3
+ "version": "0.1.52",
4
4
  "type": "module",
5
5
  "description": "CLI for the Virage RAG pipeline — provides the virage command",
6
6
  "repository": {