# almona.eg

- **Event:** [Meta Global AI Developer Hackathon](https://cerebralvalley.ai/e/meta-ai-developer-hackathon)
- **When:** Oct 26, 12:00 PM – Nov 4, 9:00 PM (EST)
- **Where:** Online
- **Team:** [Mohamed Hassan](https://cerebralvalley.ai/u/Tetos)
- **GitHub:** https://github.com/almona02/shield-os-hackathon
- **Demo video:** https://www.loom.com/share/f0ae3eb367d94de79727479913f87dd0
- **Gallery:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3348

SHIELD.OS — Tamper-Evident Approvals with Meta-Powered Risk Intelligence

Almona Co. has served aluminium workshops in Egypt since 1991 as an authorized YILMAZ dealer. We are building Almona Vertical Reality OS for machinery discovery, workshop layout, Fabricator Pro (BOM and cutting), and machine service history. Fabricator Pro is advanced in engineering (~1,539 commits) but still in pilot.

Problem: spare-part and service approvals often live in editable sheets, so after-the-fact changes weaken trust.

What we built: SHIELD.OS extracts a governance core into a clear demo:
1) immutable-log — hash-chained append-only ledger (Supabase + Postgres trigger blocking UPDATE/DELETE; API returns 405 for edits)
2) approval-chain — ordered approvals, no self-approval
3) anomaly-engine — evidence-only risk signals for human review

Demo: a technician requests Motor YLM-45 with location and evidence; SAM 3.1 scores category relevance; a manager approves; the dashboard shows chain verification, a map, and Muse Spark bilingual risk signals. We present this as a careful prototype (who / where / when / what / approved by), not a finished fraud-prevention product. Next step: Giza pilot with Fabricator Pro + SHIELD.

Live: https://shield-os-hackathon.vercel.app
GitHub: https://github.com/almona02/shield-os-hackathon
Video: https://www.loom.com/share/f0ae3eb367d94de79727479913f87dd0

## More from Meta Global AI Developer Hackathon

- [Timo](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3345)
- [Real2Simulation](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3346)
- [Hypa Intelligence](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3347)
- [Derm Forge](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3349)
- [Farfaz Learning](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3350)
- [Pinnacles](https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3351)

---

Markdown version of https://cerebralvalley.ai/e/meta-ai-developer-hackathon/hackathon/gallery/3348. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt