@contextium/cli 1.0.55 → 1.0.56
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/package.json +1 -1
- package/skills/ium/workflow.md +13 -0
package/package.json
CHANGED
package/skills/ium/workflow.md
CHANGED
|
@@ -61,6 +61,19 @@ fs.writeFileSync('CLAUDE.md', content);
|
|
|
61
61
|
" 2>/dev/null
|
|
62
62
|
```
|
|
63
63
|
|
|
64
|
+
3b. **Harness check — auto-run Initializer if needed.**
|
|
65
|
+
|
|
66
|
+
If the loaded workflow has `workflowType: harness`:
|
|
67
|
+
- Count the files in the harness state library (the library whose name contains "Harness State").
|
|
68
|
+
- If the count is ≤ 2 (only a spec/plan doc, no individual feature files), the Initializer has not run yet.
|
|
69
|
+
- In this case: **do not wait for user instruction** — immediately say:
|
|
70
|
+
|
|
71
|
+
> "This is a harness workflow with no feature files yet. Running the Initializer now to break the spec into individual features."
|
|
72
|
+
|
|
73
|
+
Then act as the Initializer agent: read the spec file from the harness state library, decompose it into individual feature files (one per DB migration, API endpoint, MCP tool, and UI component), and write each one to the harness state library using `contextium new "<harness-state-library>" -w <workspace> --stdin`. Each feature file must include: feature ID, title, layer, status (`pending`), and a detailed implementation spec with acceptance criteria.
|
|
74
|
+
|
|
75
|
+
- If the count is > 2, feature files already exist — skip this step and proceed normally.
|
|
76
|
+
|
|
64
77
|
4. Output a session context block in this exact format so all resource IDs and names are available for the rest of the session without re-querying:
|
|
65
78
|
|
|
66
79
|
```
|