signaling-kb-batched
Extract all entity types (interventions, pathways, biomarkers, conflicts) from Google Drive transcript analyses using a three-stage LLM pipeline with Claude Opus 4.5 for extraction and Haiku 4.5 for intelligent merging.
npx skills add majiayu000/claude-skill-registry --skill signaling-kb-batched --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Signaling KB Batched - Full Entity Extraction Pipeline Extract **all entity types** from transcript analyses with: - **4 entity types**: Interventions, Pathways, Biomarkers, Conflicts - **LLM-based semantic deduplication** (Opus 4.5) - resolves aliases like "Sirolimus" = "Rapamycin" - **LLM-based intelligent merging** (Haiku 4.5) - merges new data into existing entities - **Google Drive sync** with processing tracker - **Source-agnostic output** (no document titles) - **PMID/DOI citations only** (invalid citations filtered) - **Resume capability** via tracker ## Architecture ``` ┌─────────────────────────────────────────────────────────────────────────────┐ │ THREE-STAGE LLM PIPELINE │ │ │ │ ┌──────────────────┐ ┌─────────────────────┐ ┌──────────────────────┐ │ │ │ EXTRACTION │──▶│ SEMANTIC DEDUP │──▶│ LLM MERGE │ │ │ │ Claude Opus 4.5 │ │ Claude Opus 4.5 │ │ Claude Haiku 4.5 │ │ │ │ (64k tokens) │ │ (64k tokens) │ │ (64k tokens) │ │ │ └──────────────────┘ └─────────────────────┘ └──────────────────────┘ │ │ │ │ Stage 1: Extract all Stage 2: Map extracted Stage 3: Intelligently │ │ entity types from entities to existing merge new data into │ │ transcript KB entries (aliases) exi
- Architecture
- Quick Start
- Entity Types Extracted
- 1. Interventions
- 2. Pathways
- 3. Biomarkers
- 4. Conflicts
- LLM Pipeline Details
- Stage 1: Extraction (Opus 4.5)
- Stage 2: Semantic Entity Resolution (Opus 4.5)
- Stage 3: LLM Merge (Haiku 4.5)
- Cost Tracking
- API Pricing (January 2026 via OpenRouter)
- Typical Costs
python3 scripts/sync_gdrive_transcripts.py \ python3 scripts/extract_kb_openrouter.py --status nohup python3 scripts/extract_kb_openrouter.py --from-tracker > extraction.log 2>&1 & Start the monitor (checks every 10 minutes, auto-restarts if stopped) nohup bash scripts/monitor_extraction.sh > monitor.log 2>&1 & Check monitor status tail -f monitor.log Check extraction progress
What does the signaling-kb-batched skill do?
Extract all entity types (interventions, pathways, biomarkers, conflicts) from Google Drive transcript analyses using a three-stage LLM pipeline with Claude Opus 4.5 for extraction and Haiku 4.5 for intelligent merging.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill signaling-kb-batched --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.