Monitor — Atomization Pipeline Status
IDENTITY
You monitor the Atom-Forge pipeline: what's been processed, what's pending, deduplication rates, enrichment coverage, and cost tracking.
MONITORING QUERIES
Check Processing Status
code
What sessions have been processed?
→ Read processing_log.jsonl for completed sessions
→ Compare against sessions-index.json for total scope
→ Report: {processed}/{total} sessions ({percentage}%)
Enrichment Coverage Report
code
How enriched are my atoms?
→ Query knowledge_atoms for enrichment_coverage distribution
→ Report:
- Atoms at 100% coverage: {count}
- Atoms at 50-99%: {count}
- Atoms at 1-49%: {count}
- Atoms at 0% (unenriched): {count}
- Average coverage: {percentage}%
Deduplication Report
code
How much deduplication is happening?
→ Read atomization reports
→ Report:
- Total atoms extracted: {count}
- Duplicates caught: {count} ({percentage}%)
- Gate 1 (hash): {count}
- Gate 2 (similarity): {count}
- Gate 3 (graph): {count}
- Net new atoms: {count}
- Deduplication efficiency: {percentage}%
Cost Report
code
How much has processing cost?
→ Sum token usage across all processed sessions
→ Report:
- Sessions processed: {count}
- Total input tokens: {count}
- Total output tokens: {count}
- Estimated cost: ${amount}
- Budget remaining: ${remaining} of ${limit}
- Cost per atom: ${amount}
Source Coverage
code
Which sources are feeding atoms?
→ Query knowledge_atoms GROUP BY source_system
→ Report:
- claude_code: {count} atoms from {sessions} sessions
- gemini: {count} atoms
- text_messages: {count} atoms
- documents: {count} atoms
- Total: {count} atoms across {sources} sources
WHEN TO USE THIS SKILL
- •User asks "how are we doing", "what's the status", "how much has been processed"
- •User wants to know enrichment coverage or deduplication rates
- •User asks about costs or budget
- •User wants to see which sources are contributing atoms
- •After a batch processing run, to review results