# Figment Submission Checklist

Status: living checklist for evidence. Keep claims in README, submission copy, demo video, and social posts aligned with this file.

Primary tracker: [adversarial review action items](adversarial-review-action-items.md).

## Required Submission Artifacts

| Artifact | Status | Link / evidence |
| -------- | ------ | --------------- |
| Public Hugging Face Space | Target exists; runnable proof needed | https://huggingface.co/spaces/build-small-hackathon/figment |
| Space cold boot with app files present | Not verified | Last known public API state: `runtime.stage=NO_APP_FILE` with only metadata files present |
| Demo video | Proof needed | Pending |
| Social post | Proof needed | Pending |
| Safety statement | Present | [safety_statement.md](safety_statement.md) |
| User test notes | Template present; results needed | [user_test_notes.md](user_test_notes.md) |
| License | Present | [../LICENSE](../LICENSE) |
| Live hosted Omni trace | Eval traces present; final demo trace still needed | Baseline: `traces/hosted_omni_eval_20260607T194833Z.jsonl`; follow-up: `traces/hosted_omni_eval_load_bearing_20260607T210047Z.jsonl` |
| No-cloud/off-grid trace | Proof needed before claiming Off the Grid achieved | Pending |
| Hosted Omni eval results | Measured | [hosted_omni_eval_results.md](hosted_omni_eval_results.md): 31/50 whole-output competence, 8/50 full fallback, 480/650 model-retained fields, 170/650 deterministic patches, 50/50 final validation in the follow-up run |
| Local 4B + Parakeet eval results | Proof needed | Pending local no-cloud 50-case eval and ASR proof |
| Parameter/evidence ledger | Present; organizer confirmation pending | [model_parameter_evidence_ledger.md](model_parameter_evidence_ledger.md) |

## Badge And Claim Status

| Claim / badge area | Submission wording allowed now | Evidence needed to upgrade |
| ------------------ | ------------------------------ | -------------------------- |
| Backyard AI | Targeted; built for a real trained responder, with identity withheld for privacy | Completed user-test notes from that responder on synthetic or de-identified scenarios. Do not claim use, testing, validation, approval, or endorsement before notes exist |
| Off the Grid | Targeted / proof-needed | Recorded no-cloud run using self-hosted Omni on adequate local hardware or a smaller verified local stack. Hosted API evidence does not count |
| Hosted Gradio Space | Targeted / proof-needed; not claimed runnable | Public Space app files present, cold boot, typed intake run, and route/fallback trace. Last known state is `NO_APP_FILE` |
| Demo video | Targeted / proof-needed | Final video link showing only verified routes and labeling fallbacks honestly |
| Social post | Targeted / proof-needed | Final social post link with achieved-versus-targeted wording |
| Llama Champion | Targeted / proof-needed | Eligible local model route through llama.cpp with trace or eval evidence |
| Sharing is Caring | Targeted / proof-needed | Public Space, repo, demo video, social post, and open trace links |
| Well-Tuned | Stretch only / proof-needed | Published fine-tuned model or adapter used by the app, plus measured result. Fallback output cannot count |
| Field Notes | Tentative | Organizer confirmation and final field-note artifact |
| Off-Brand | Targeted / proof-needed | Final demo or social artifact meeting organizer criteria |

## Off-Grid Evidence Boundary

Omni is not architecturally disqualified from Off the Grid: it can support the claim if self-hosted on adequate local hardware with no runtime cloud APIs. The current repo does not yet include that recorded proof. Until a no-cloud run exists, use targeted or proof-needed language.

## User-Test Evidence Boundary

The README may say the project is built for a real trained responder. It should not say that responder used, approved, validated, or endorsed Figment until [user_test_notes.md](user_test_notes.md) contains factual session notes.

## Eval Evidence Boundary

The hosted Omni eval proves model and fallback behavior through the eval harness, not a public Space cold boot. Use the follow-up run as the current hosted eval score: 31/50 whole-output hosted competence, 8/50 full fallback, 480/650 model-retained fields, 170/650 deterministic patches, and 50/50 final validation.

Final validation is app safety. Whole-output model competence and field-level model retention are the model-load-bearing metrics. Deterministic fallback and deterministic patches must stay visible in traces, scorecards, submission copy, and the demo.

## Submission Copy Boundaries

- Space: may say the target Space exists; do not say it is runnable until app files, cold boot, typed intake, and trace labeling are verified from the public Space URL.
- Demo video and social post: use pending placeholders until final links exist.
- Backyard AI: may say built for a real trained responder; do not say the target user used or tested Figment until factual notes exist.
- Off the Grid: claim only after a recorded no-cloud run.
- Llama Champion: claim only after an eligible llama.cpp route runs with trace or eval evidence.
- Well-Tuned: claim only after a published fine-tuned model or adapter is used by the app and measured.
- Parameter compliance: cite the [model parameter/evidence ledger](model_parameter_evidence_ledger.md), including the Omni 31B body-count versus 33B sidebar ambiguity and organizer-confirmation status.