minutes-ingest

1# /minutes-ingest
2 
3Process meetings through the knowledge extraction pipeline to update person profiles, append to the knowledge log, and maintain the index.
4 
5## Prerequisites
6 
7The `[knowledge]` section must be configured in `~/.config/minutes/config.toml`:
8 
9```toml
10[knowledge]
11enabled = true
12path = "/path/to/knowledge/base"
13adapter = "wiki"  # or "para", "obsidian"
14engine = "none"   # or "agent" for LLM extraction
15min_confidence = "strong"
16```
17 
18If not configured, explain what's needed and offer to help set it up.
19 
20## How to run
21 
22### Single meeting
23```bash
24minutes ingest ~/meetings/2026-04-03-strategy-call.md
25```
26 
27### All normal meetings (backfill)
28```bash
29minutes ingest --all
30```
31 
32### Preview without writing (recommended first time)
33```bash
34minutes ingest --all --dry-run
35```
36 
37## What it does
38 
391. **Reads** each meeting's YAML frontmatter (decisions, action_items, entities, intents)
402. **Extracts** structured facts with confidence levels and source provenance
413. **Updates** person profiles in the knowledge base (adapter-dependent format)
424. **Appends** to `log.md` with a timestamped entry for each ingested meeting
435. **Skips** facts that already exist (deduplication) or are below the confidence threshold
446. **Excludes** meetings designated `sensitivity: restricted` from automated knowledge-base ingestion
45 
46## Safety guarantees
47 
48- **`engine = "none"` (default)**: Only extracts from parsed YAML frontmatter. No LLM involved, zero hallucination risk.
49…