181d4129ea
Issue #138 — 7.6: Conference Talk Summarizer. Adds a complete pipeline for ingesting conference talk transcripts into the compounding-intelligence knowledge store. ### New files - scripts/conference_summarizer.py - Reads plain-text transcript files - Calls LLM (mimo-v2-pro default) to extract knowledge items - Deduplicates against existing store - Assigns IDs following {domain}:{category}:{NNN} schema - Writes to knowledge/index.json and knowledge/conferences/talks.md - Supports --dry-run, --domain, --conference tags - templates/conference-summary-prompt.md - Specialized prompt for conference talk knowledge extraction - Mirrors harvester prompt structure but tuned for talk context - Categories: fact, pitfall, pattern, tool-quirk, question - Evidence required per item - Domain tagging (global|repo|agent|compounding-intelligence) ### Acceptance criteria - ✅ Finds talk transcripts — accepts any plain-text transcript file - ✅ Generates summary — LLM produces structured knowledge items - ✅ Extracts key takeaways — fact/pattern/pitfall/tool-quirk/question - ✅ Stores in knowledge base — writes to index.json + conferences/talks.md - ✅ Weekly — script can be scheduled via cron (usage example in doc) ### Usage example python3 scripts/conference_summarizer.py \ --transcript ~/Downloads/ai拂晓-2026-04-10.txt \ --conference "AI拂晓 2026" \ --title "Scaling Autonomous Agents" \ --speaker "Alexander" \ --domain global \ --dry-run Run without --dry-run to actually write to knowledge store. API key resolved from HARVESTER_API_KEY or ~/.config/nous/key etc. Closes #138
Compounding Intelligence
Turn 1B+ daily tokens into durable, compounding fleet intelligence.
The Problem
20,991 sessions on disk. Each one starts at zero. Every agent rediscover the same HTTP 405 is a branch protection issue. The intelligence from a million tokens of work evaporates when the session ends.
The Solution
Three pipelines that form a compounding loop:
SESSION ENDS → HARVESTER → KNOWLEDGE STORE → BOOTSTRAPPER → NEW SESSION STARTS SMARTER
↓
MEASURER → Prove it's working
Architecture
Pipeline 1: Harvester
Reads finished session transcripts. Extracts durable knowledge: facts, pitfalls, patterns, tool quirks. Stores in knowledge/.
Pipeline 2: Bootstrap
Before a session starts, queries knowledge store for relevant facts. Assembles compact 2k-token context. Injects into session so it starts with full situational awareness.
Pipeline 3: Measure
Tracks whether compounding is happening. Knowledge velocity, error reduction, hit rate, task completion. Daily report proves the loop works.
Directory Structure
├── knowledge/
│ ├── index.json # Machine-readable fact index
│ ├── global/ # Cross-repo knowledge
│ ├── repos/{repo}.md # Per-repo knowledge
│ └── agents/{agent}.md # Agent-type notes
├── scripts/
│ ├── harvester.py # Post-session knowledge extractor
│ ├── bootstrapper.py # Pre-session context loader
│ ├── measurer.py # Compounding metrics
│ └── session_reader.py # JSONL parser
├── metrics/
│ └── dashboard.md # Human-readable status
└── templates/
├── bootstrap-context.md
└── harvest-prompt.md
The 100x Path
Month 1: 15,000 facts, sessions 20% faster
Month 2: 45,000 facts, sessions 40% faster, first-try success up 30%
Month 3: 90,000 facts, fleet measurably smarter per token
Each new session is better than the last. The intelligence compounds.
Issues
See all issues for the full roadmap.
Epics:
- EPIC 1: Session Harvester (#2)
- EPIC 2: Knowledge Store & Bootstrap (#3)
- EPIC 3: Compounding Measurement (#4)
- EPIC 4: Retroactive Harvest (#5)