@skilljack/mcp 0.5.0 → 0.5.1
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.
- package/README.md +24 -3
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -323,14 +323,35 @@ Each skill subdirectory must contain a `SKILL.md` file with YAML frontmatter inc
|
|
|
323
323
|
|
|
324
324
|
## Testing
|
|
325
325
|
|
|
326
|
+
### Manual Testing with MCP Inspector
|
|
327
|
+
|
|
326
328
|
```bash
|
|
327
|
-
# Build first
|
|
328
329
|
npm run build
|
|
330
|
+
npm run inspector -- /path/to/skills
|
|
331
|
+
```
|
|
332
|
+
|
|
333
|
+
### Automated Evals (Development Only)
|
|
334
|
+
|
|
335
|
+
The `evals/` directory contains an evaluation framework for testing skill activation across different delivery modes. Evals are only available when developing from source (not included in the npm package).
|
|
336
|
+
|
|
337
|
+
```bash
|
|
338
|
+
# Clone the repo first
|
|
339
|
+
git clone https://github.com/olaservo/skilljack-mcp.git
|
|
340
|
+
cd skilljack-mcp
|
|
341
|
+
|
|
342
|
+
# Install dev dependencies (includes claude-agent-sdk for evals)
|
|
343
|
+
npm install
|
|
329
344
|
|
|
330
|
-
#
|
|
331
|
-
|
|
345
|
+
# Build and run evals
|
|
346
|
+
npm run build
|
|
347
|
+
npm run eval # Default: greeting task, MCP mode
|
|
348
|
+
npm run eval -- --task=xlsx-openpyxl # Specific task
|
|
349
|
+
npm run eval -- --mode=native # Native skill mode
|
|
350
|
+
npm run eval -- --mode=mcp+native # Both MCP and native enabled
|
|
332
351
|
```
|
|
333
352
|
|
|
353
|
+
See [evals/README.md](evals/README.md) for details on available tasks, modes, and findings about activation behavior differences.
|
|
354
|
+
|
|
334
355
|
## Related
|
|
335
356
|
|
|
336
357
|
- [Agent Skills Specification](https://agentskills.io)
|