Converts user-provided images into validated, decision-ready action drafts. Routes images by inferred intent across domains: events from flyers, macros from meal photos, places to save, products to price, receipts to log, documents to file, civic issues to report. Supports reverse image search via Yandex and Google. NOT for generic OCR, computer vision research, or surveillance.
Skill name: ocas-look
Version: 2.5.2
Type:
Layer: ocas-look
Author:
Converts user-provided images into validated, decision-ready action drafts. Routes images by inferred intent across domains: events from flyers, macros from meal photos, places to save, products to price, receipts to log, documents to file, civic issues to report. Supports reverse image search via Yandex and Google. NOT for generic OCR, computer vision research, or surveillance.
look.ingest.image— ingest image(s) with optional EXIF and device pre-parselook.propose.actions— generate ActionDrafts with DecisionRecordslook.execute.action— execute a confirmed draft (requires explicit approval)look.rollback.action— attempt rollback for reversible actionslook.status— last ingest, pending drafts, items awaiting confirmationlook.config.set— update configurationlook.journal— write journal for the current run; called at end of every runlook.update— pull latest from GitHub source; preserves journals and datalookup.reverse_search— perform Google reverse image search on an image URL or local file. Returns matching pages, titles, and similar image URLs.
See SKILL.md for outputs, journals, and persistence rules.
| File | Purpose |
|---|---|
SKILL.md |
Skill definition |
references/ |
Supporting documentation |
scripts/ |
Helper scripts |
- [2.4.5] - 2026-04-26
- Changed
- [2026-04-04] Spec Compliance Update
- Changes
- Validation
- [2.4.1] - 2026-04-08
- Storage Architecture Update
- [2.4.0] - 2026-04-08
Read SKILL.md for operational details, schemas, and validation rules.
Read references/ for detailed specifications and examples.
MIT License — see LICENSE for details.