Cup
Manually trigger CocoCupper analysis of the current session's work.
Install
npx skills add Snowflake-Labs/cocoplus --skill cup -g -a claude-code -y- Repository
- Snowflake-Labs/cocoplus727
- Category
- Engineering
- Updated
- 1mo ago
- License
- MIT
- Context cost
- Lightabout 369 tokens
- Runs scripts
- No, instructions only
What it does
Manually trigger CocoCupper analysis of the current session's work. Runs immediately (not in background) and writes findings to .cocoplus/grove/cupper-findings.md.
Your objective is to manually trigger CocoCupper analysis.
Before proceeding, verify that .cocoplus/ exists.
If not: output "CocoPlus not initialized in this directory. Run $pod init to begin." Then stop.
Activate the coco-cupper subagent with the following context:
- Current lifecycle artifacts in
.cocoplus/lifecycle/ - Current flow.json and stage statuses
- Any quality findings files matching
.cocoplus/quality-findings-*.md - Current memory/decisions.md
The coco-cupper agent will analyze and write findings to .cocoplus/grove/cupper-findings.md.
Output: "CocoCupper analysis complete. Findings written to .cocoplus/grove/cupper-findings.md. Run $cup history to review, or $patterns promote [finding-id] to promote a finding to CocoGrove."
Anti-Rationalization
| Shortcut / Temptation | Why It Fails |
|---|---|
| Run analysis without loading quality findings and memory context | CocoCupper's value comes from cross-referencing lifecycle artifacts, incomplete context produces shallow findings |
| Append to findings file without a unique finding ID | Duplicate or ID-less findings cannot be referenced by $patterns promote and become orphaned |
Exit Criteria
-
.cocoplus/grove/cupper-findings.mdhas been updated with at least one new## Findingsection - Each new finding has a unique ID, Type, Severity, and Date field
- Output confirms analysis completed with a path to findings file
