Add prominent warnings and explicit instructions to prevent agents from
creating individual markdown files instead of using the MCP catalog.
Changes:
- Add "CRITICAL: MCP CATALOG USAGE" section at the top of the prompt
- Explicitly prohibit Write tool for individual agent discoveries
- Specify that ONLY Round 4 final report should be written to file
- Add detailed catalog_upsert usage examples
- Update all round descriptions with CRITICAL catalog usage instructions
- Add "IMPORTANT - Catalog Usage Rules" section
Key Instructions Added:
- "DO NOT use the Write tool to create separate markdown files"
- "ALL agent findings MUST be written to MCP catalog using catalog_upsert"
- "ONLY Round 4 Final Synthesis writes to local file using Write tool"
- "DO NOT use Write tool for individual agent discoveries in Rounds 1-3"
This ensures agents store findings in the MCP catalog for cross-agent
collaboration, with only the final consolidated report written to disk.
Perform comprehensive database discovery through 6 collaborating subagents using ONLY MCP server tools (`mcp__proxysql-stdio__*`). Output: Single comprehensive markdown report.
## ⚠️ CRITICAL: MCP CATALOG USAGE
**ALL agent findings MUST be stored in the MCP catalog using `catalog_upsert`.**
**DO NOT use the Write tool to create separate markdown files for individual agent discoveries.**
- Round 1-3 findings: Use `catalog_upsert` ONLY
- Round 4 final report: Use both `catalog_upsert` AND Write tool (for the single consolidated report)
- Round 5 meta analysis: Use `catalog_upsert` ONLY
**WRONG:** Using Write tool for each agent's findings creates multiple markdown files
**RIGHT:** All findings go to MCP catalog, only final report is written to file
Example correct usage:
```python
# After discovery, write to catalog
catalog_upsert(
kind="structural", # or statistical, semantic, query, security, meta_analysis, question_catalog
key="round1_discovery",
document="## Findings in markdown..."
)
```
Only in Round 4 Final Synthesis:
```python
# Write the consolidated report to catalog AND file
6. **CONSTRUCTIVE META**: META agent provides actionable, specific improvements
7. **QUESTION CATALOGS**: Each agent MUST generate a question catalog with executable answer plans
2. **CATALOG FOR FINDINGS**: ALL agent findings MUST be written to MCP catalog using `catalog_upsert` - NEVER use Write tool for individual agent discoveries
3. **NO INTERMEDIATE FILES**: DO NOT create separate markdown files for each agent's findings - only the final synthesis should be written to a local file
4. **EVIDENCE-BASED**: All claims backed by database evidence
5. **SPECIFIC RECOMMENDATIONS**: Provide exact SQL for all changes
6. **QUANTIFIED IMPACT**: Include expected improvements with numbers
Rene Cannao
3 months ago