메뉴
Daily cron that scans IMAGE_SOURCE_DIR, imports new images (jpg/png/heic/webp/gif) into gbrain via importImageFile, embeds them through Voyage multimodal-3 (1024d), and writes them as pages with cross-modal search capability (gbrain search --image).
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Classify vector-similar pages against each orphan page via Haiku 4.5 and add typed inbound references (link_type="related", relation encoded in context). Goal: chip away at the 1803/1952 orphan pile (v1 wiki flat import) across multiple sweep runs. Default is --dry-run; --apply does DB link inserts + markdown sentinel upsert on candidate pages that exist on disk, then commits ~/brain changes.
Merge WITHIN-SOURCE duplicate entity-variant pages (e.g. people/lucien + people/lucien-chen, companies/tp-link-system-inc + tp-link-systems-inc) into one canonical node: repoint links + facts + display-name aliases, register a slug redirect, soft-delete the alias pages. An LLM disambiguates each cluster (merge / ambiguous / distinct); ambiguous first-name-vs-multiple-full-names cases are quarantined, never merged. Default --dry (simulate + report, zero writes). Cross-source pages are NEVER touched — the two-axis design keeps them legitimately separate.
One-shot sweep that reads every page in the brain, extracts every person/company/concept/project mentioned via Haiku NER, dedupes, tiers by mention count + context, and creates stub pages for entities that lack one. Tier 2 candidates get Tavily web-search augmentation; Tier 3 stays brain-only (confidence: low). This is the primary G1 payoff skill from JARVIS-NEXT-STEPS.md §5.
Daily sweep: runs kos-lint + staleness detection + knowledge-gap analysis, then produces a MEMORY-format digest ready to be appended to OpenClaw's MEMORY.md by the digest-to-memory skill (Sun 22:00 weekly). Also writes a dashboard to ~/brain/agent/dashboards/knowledge-health-<date>.md.
Bridge an upstream gap: extract_atoms never stamps `frontmatter.concepts` on page-derived atoms, but synthesize_concepts consumes ONLY that field — so concept synthesis is permanently skipped ("no atoms with concept refs"). This script batch-labels atoms lacking the field with 1-3 kebab-case shared-vocabulary topic labels (sonnet) and stamps them, so upstream synthesize_concepts can tier + materialize concepts/<label> pages. Idempotent: only touches atoms WITHOUT `concepts`; safe to rerun after each extract_atoms drain. Retires when the upstream fix lands.
One-command bulk-ingest pipeline (D→I→K→W) for a single isolated source. Sequences the 7 fork stages — register → ingest → embed → enrich → graph → (per-entity synth, opt-in) → corpus-synth — over a folder of frontmatter'd .md files, each idempotent/incremental so re-runs only do new work. Turns a systematic corpus (product docs, FAQ, a research library) into a queryable knowledge asset: facts (RAG) + entities + graph + corpus-level viewpoints.
| name | image-ingest |
| description | Daily cron that scans IMAGE_SOURCE_DIR, imports new images (jpg/png/heic/webp/gif) into gbrain via importImageFile, embeds them through Voyage multimodal-3 (1024d), and writes them as pages with cross-modal search capability (gbrain search --image). |
| triggers | ["ingest images","image ingest","scan image dir"] |
| status | scaffolded — needs VOYAGE_API_KEY + IMAGE_SOURCE_DIR set before bootstrap |
Wraps gbrain import <IMAGE_SOURCE_DIR> for the v0.36.6.0 cross-modal
search wave. Routed to importImageFile when GBRAIN_EMBEDDING_MULTIMODAL=true
kind=source + embedding_multimodal vector(1024)
populated via Voyage multimodal-3.Register Voyage account + get API key at https://www.voyageai.com. Free tier covers ~50M tokens; multimodal-3 is $0.05/M tokens. Pilot batch of ~500 images ≈ $0.05.
Curate IMAGE_SOURCE_DIR. Examples:
~/jarvis-images/ — manually drop screenshots / charts / photos
worth indexing~/Downloads/notion-r5a-capture/ — already on disk from prior export~/Pictures/ if it points to Photos.app library — those are not
individually addressable)Fill plist template at
scripts/launchd/com.jarvis.image-ingest.plist.template:
<FILL:VOYAGE_API_KEY> with real key<FILL:IMAGE_SOURCE_DIR> with absolute path~/Library/LaunchAgents/com.jarvis.image-ingest.plistBootstrap:
launchctl bootstrap gui/$UID ~/Library/LaunchAgents/com.jarvis.image-ingest.plist
First manual kick:
launchctl kickstart -k gui/$UID/com.jarvis.image-ingest
tail -50 skills/kos-jarvis/image-ingest/image-ingest.stdout.log
bun run skills/kos-jarvis/image-ingest/run.tsIMAGE_SOURCE_DIR recursively, dispatches each
image extension via gbrain import <dir> (gbrain's walker auto-detects
image extensions when GBRAIN_EMBEDDING_MULTIMODAL=true). Skips files
already imported (gbrain's idempotency via content_hash).~/brain/.agent/image-ingest/<ISO>.json records
imported count + cost estimate. latest.json symlink for /status.Voyage multimodal-3 at $0.05/M tokens. Per-image typical: ~300-800 tokens
(image patches + caption tokenization). 100 image batch ≈ $0.015. Cron
log line includes cost_usd_estimate.
If monthly spend > $5 unexpectedly: check ~/brain/.agent/image-ingest/latest.json
for imported_count (Voyage quotas per-account so spend can't run away
silently).
Verify multimodal column populated:
psql -h 127.0.0.1 -U chenyuanquan -d gbrain -c \
"SELECT COUNT(*) FROM content_chunks WHERE embedding_multimodal IS NOT NULL;"
Test cross-modal query:
bin/gbrain search "hackathon photos" 2>&1 | head
bin/gbrain query "what charts did I capture from acme corp?" 2>&1 | head
gbrain reindex --multimodal ad-hocScaffolded 2026-05-19 post v0.37.0.0 sync (per TODO L177 P1 + JARVIS-ARCHITECTURE §6.29 Followup F decision (a) "Full enable + cron"). Awaits VOYAGE_API_KEY + IMAGE_SOURCE_DIR fill before bootstrap.