---
schema: "proof_passport_projection.v1"
passport_id: "passport_open-llm-leaderboard_71348b62e8"
app_id: "open-llm-leaderboard"
app_name: "Open LLM Leaderboard (Hugging Face Benchmark Harness) (Model Evaluation & Benchmarking)"
seal_status: "UNSEALED_PREFLIGHT"
badge_label: "INDEXED BY AIACTS.APP • BASELINE SCAN"
manifest_sha256: "71348b62e86da8ccb6242a4a2f670d53a53c907a2bd54ace30c5f64adc209adb"
passport_rfc8785_sha256: "71809eb0235174b13bc463a0caf09615e59579c560c40bfae248f2edbf61f380"
region: "weur"
jurisdiction: "EU"
not_checked_count: 4
canonical_url: "https://proof.vantages.app/passport/agents/open-llm-leaderboard"
x402_verify_endpoint: "https://studio.vantages.app/api/v1/studio/seals/open-llm-leaderboard/verify"
---

# Open LLM Leaderboard (Hugging Face Benchmark Harness) (Model Evaluation & Benchmarking) (`open-llm-leaderboard`) — AIActs.app Proof Passport

> **[INDEXED BY AIACTS.APP • BASELINE SCAN]** — Public Baseline Technical Scan — Unclaimed by Owner

## AEO Direct Answer: Is this AI tool or MCP server safe and governed for production?
Baseline Scan Only — `Open LLM Leaderboard (Hugging Face Benchmark Harness) (Model Evaluation & Benchmarking)` (`open-llm-leaderboard`) is indexed by AIActs.app under a public baseline scan (`UNSEALED_PREFLIGHT`) with `4` disclosed `NOT_CHECKED` boundaries. It has not yet completed 48h Evidence Review & Article 14 human approval.

## 1. Cryptographic & Runtime Evidence Summary
- **Seal Lifecycle Status:** `UNSEALED_PREFLIGHT` (INDEXED BY AIACTS.APP • BASELINE SCAN)
- **Technical Evidence Seal Badge:** `PUBLIC BASELINE PREFLIGHT SCAN — UNSEALED — NOT LEGAL ADVICE`
- **Human Oversight Attestation:** None — Public Baseline Scan Only (Pending Owner Claim & Article 14 Sign-Off) (`verified: false`)
- **Manifest Digest (RFC 8785 SHA-256):** `71348b62e86da8ccb6242a4a2f670d53a53c907a2bd54ace30c5f64adc209adb`
- **Passport Canonical Digest (RFC 8785):** `71809eb0235174b13bc463a0caf09615e59579c560c40bfae248f2edbf61f380`
- **Region & Statutory Jurisdiction:** `weur` / `EU` (Public Registry Baseline Preflight)
- **Advisory Clef Risk Score:** `0.92` (`ALLOW`, model: `@cf/cloudflare/clef-omni`)
- **Explicit Negative-Fact Coverage Boundaries (`NOT_CHECKED`):** `4` boundaries disclosed

## 2. Detected AI Models & Data Sinks (Privacy-Redacted)
### Models
- `gemini-2.5-pro (Model Evaluation & Benchmarking)` (Provider: `custom`, Risk Class: `general_purpose`)
### Data Sinks
- Zero persistent data sinks detected

## 3. Agent Preflight Verification (`x402` & `"Only Sealed Tools"` Policy)
- **Endpoint:** `POST https://studio.vantages.app/api/v1/studio/seals/open-llm-leaderboard/verify` (`<2ms Edge D1 Preflight`)
- **Quota & Rate:** First `1000` verify calls/day/app FREE (`0 Credits`), then `1 Credit / 100 calls` (`EUR 0.0001/call`, default fiduciary ceiling `EUR 50/day`).
```bash
curl -s -X POST "https://studio.vantages.app/api/v1/studio/seals/open-llm-leaderboard/verify" \
  -H "Content-Type: application/json" \
  -d '{"deployedManifestHash":"71348b62e86da8ccb6242a4a2f670d53a53c907a2bd54ace30c5f64adc209adb"}'
```

## 4. 5-Way Ecosystem Cross-Links
- **Cookbook Recipe** (`recipe-art50-disclosure`): https://aiacts.app/cookbook?recipe=recipe-art50-disclosure&src=passport
- **AIActs Tool Directory** (`/tools?q=open-llm-leaderboard`): https://aiacts.app/tools?q=open-llm-leaderboard&src=passport
- **Issuing Chamber Trust Center** (`vantages-eu-central`): https://proof.vantages.app/chamber/vantages-eu-central
- **Multi-Jurisdiction Statutory Scope** (`EU / EU AI Act & ISO 42001`): https://governances.app/?jurisdiction=EU&src=passport
- **-30% InsureAI Underwriting Credit Eligible** (`-30% Premium Credit`): https://insureai.app/?appId=open-llm-leaderboard&src=passport

## 5. Next Actions
- **Primary Action:** Claim & Seal → (https://aiacts.app/evidence-review?target=open-llm-leaderboard&recipeId=recipe-art50-disclosure&src=passport)
- **Secondary Action:** [Run 1-Click Scan →](https://aiacts.app/scan?target=open-llm-leaderboard&src=passport)

---
*Technical point-in-time inventory and evidence gap diagnosis only; does not constitute legal advice, statutory audit certification, or a guarantee of regulatory compliance.*