Status: Public scaffold. Release status: scaffolded. License pending human review.
This repository is the public proof-control layer for Francisco Abner Rivera's Franzabner technical surface. It keeps public claims tied to live repositories, artifact status, boundary review, and human approval before those claims move into profile navigation, Hugging Face context, or draft Upwork proof routing.
The proof stack has now moved from first-batch publication planning into live navigation for:
- the first five public scaffolds;
- the first safe older mechanical/electrical/embedded repo alignment batch;
- the EPI spine research scaffold alignment batch;
- the paper-specific EPI paper/research scaffold cleanup batch.
- the RIB breadboard public-boundary cleanup as a breadboard/reference scaffold.
- the first derivative public boundary repo,
local-ai-lab-boundary. - the second derivative public repo,
dgx-ai-workstation-notes. - the third derivative public repo,
fine-tuning-lab-notes.
Higher-risk rack, agent, Fusion, KiCad, production RIB, and production-oriented repos remain held for boundary review or derivative-only planning. local-ai-lab-boundary provides the public-safe boundary model that replaces pressure to expose real studio, lab, dotfiles, endpoint, credential, private prompt, private corpus, private weight, or private harness material. dgx-ai-workstation-notes adds the public-safe DGX/NVIDIA workstation workflow-notes layer without exposing real topology, endpoints, credentials, tokens, private corpora, private weights, private prompts, private agent harnesses, or sealed YOSO-YAi IP. fine-tuning-lab-notes adds the public-safe fine-tuning documentation layer without exposing private corpora, private weights, training logs, eval results, model outputs, hosted artifacts, or Hugging Face releases.
| Control | Purpose |
|---|---|
proof-matrix.md |
Maps proof lanes to repos, artifacts, status, and review needs. |
claim-register.md |
Records allowed, draft, and forbidden claims. |
CLAIMS.md |
Claim-review summary of allowed, review-required, and forbidden claim classes. |
artifact-status.md |
Tracks live scaffold, merged scaffold, draft, held, and human-review-required labels without implying release. |
review-log.md |
Records human review decisions before public use. |
REVIEW_CHECKLIST.md |
Gate before the proof stack is used for external proof references. |
REPO_MAP.md |
Maps Franzabner repos by proof lane and review gate. |
WAVE_PLAN.md |
Shows staged rollout across navigation, hardware, agents, EPI, AI/HF, Omniverse, and monetization routing. |
HUGGINGFACE_MAP.md |
Keeps Hugging Face as a release surface only. |
UPWORK_MAP.md |
Keeps Upwork language as draft proof routing only. |
MONETIZATION_MAP.md |
Records monetization-readiness boundaries without live offers, pricing, or client claims. |
BOUNDARY_MAP.md |
Summarizes public, private, and sealed exclusions by technical lane. |
LICENSE_DECISION.md |
Records that license selection is still pending human review. |
| Group | Repos | Status |
|---|---|---|
| First public scaffolds | franzabner-proof-stack, public-pcb-review-checklist, engineering-agent-review-gates, hf-card-templates, technical-readme-lab |
Live public scaffolds. Release status: scaffolded. License pending human review. |
| Derivative public boundary repos | local-ai-lab-boundary, dgx-ai-workstation-notes, fine-tuning-lab-notes |
Live public scaffolds. local-ai-lab-boundary is the local AI lab boundary model. dgx-ai-workstation-notes is public-safe DGX/NVIDIA/workstation workflow documentation. fine-tuning-lab-notes is public-safe fine-tuning lab-note documentation with synthetic-only examples, dataset prep boundaries, model release gates, evaluation note templates, and Hugging Face planned/template-only routing. No real topology, endpoints, credentials, tokens, private corpora, private weights, private prompts, private agent harnesses, sealed YOSO-YAi IP, model, dataset, Space, training run, training log, model output, eval result, hosted artifact, or HF release claim. |
| Mechanical / electrical / embedded old-repo batch | electrical-panel-dressing, field-to-architect, mission-critical-discipline, pico2-pio-i2c-reference |
Public-surface alignment merged. Still scaffold/proof surfaces, not release evidence. |
| RIB breadboard reference | rib-breadboard |
Public-boundary cleanup merged. Public breadboard/reference scaffold only; production RIB implementation remains sealed. No production board files, firmware internals, BOMs, Gerbers, pin maps, routing, manufacturing approval, certification, sign-off, or production-ready claim. |
| EPI spine | energy-per-intelligence, epi-bench, epi-meter |
Public-surface alignment merged. Research/tooling/hardware measurement scaffolds only. |
| Paper-specific EPI | attention-head-surgery-epi, mixed-quant-epi, expert-pruning-epi |
Public-surface alignment merged. Paper/research scaffolds only; no released method, benchmark, dataset, model, Space, eval result, deployment, validated method, or proven result claim. |
| Held group | Repos or lane | Reason |
|---|---|---|
| Risky production-adjacent repos | rack, agent, Fusion, KiCad production, production RIB, private infrastructure lanes | Need boundary review or derivative-only planning before linking as proof. Fine-tuning lab notes now route through fine-tuning-lab-notes; any additional fine-tuning or evaluation derivative remains held until review. |
| Hugging Face artifacts | models, datasets, Spaces, reports, hosted artifacts | No artifact claim without evidence and human approval. |
| Upwork routes | service copy, rates, delivery promises, client outcomes | Draft proof routing only. |
No claim is clean public proof until it has:
- an evidence row;
- an artifact status;
- a public/private/sealed boundary note;
- a review requirement;
- a refusal rule for unsupported or unsafe claims.
This scaffold does not authorize additional GitHub repo creation, Hugging Face artifact creation, live Upwork copy, license selection, release claims, client claims, deployment claims, manufacturing approval, certification, sign-off, production-ready status, model weights, datasets, Spaces, benchmarks, eval results, or hosted artifact claims.