A proof-of-work builder community for ML, Deep Learning, DSA, MLOps, AI systems, replay/evidence systems, and AI governance.
Build. Test. Commit. Receipt. Repeat.
No commit. No claim. No receipt. No authority.
Most groups reward talking. We require receipts.
Learn → Build → Test → Commit → Receipt → Review → Improve
- No fake progress
- No hype without evidence
- Claims require receipts
- Failed experiments are welcome when documented
- Builders only
- ML Foundations
- Deep Learning
- DSA
- Weaver Systems (receipt ledgers, claim registries, replay tools, governance systems)
- Join our Discord (replace with actual link)
- Read the rules and templates in this repo
- Introduce yourself and post your first receipt
This GitHub repository is the official evidence layer for Weaver Forge.
Build. Test. Commit. Receipt. Repeat.
Check that every file in receipts/ includes the required sections and a Commit: line:
python scripts/validate_receipts.pyExit code 0 means all receipts passed; 1 means at least one receipt is missing required fields.
| Package | Status |
|---|---|
| Grok Build narrow clean rebuild Witness package | NOT READY — package version 1.0.0-rc4; fixed immutable annotated tag grok-build-witness-v1.0.0-rc4 → commit 039b46737c5968a81fb756d7a6d1d0dd57b6ad96; rc4 static blind audit COMPLETE with final disposition NOT READY (40 integrated blockers); Phase 0 audit intake COMPLETE; Phase 1 documentation remediation on main; technical implementation remediation NOT YET BEGUN; main prepared toward possible future rc5 candidate (rc5 tag does not exist); Independent Witness reproduction NOT PERFORMED; Independent Witness PASS NONE; C-014 NOT_STARTED; overall PARTIAL |
See REPRODUCE.md for clone prerequisites, validation commands, GitHub Actions checks, and witness review steps.