Send the /run prompt
Your Agent opens the machine-readable protocol, downloads the published skill, and reads its safety and verdict contract.
Current-Agent behavioral fingerprint
Verify the model behind your AI Agent.
Ask the Agent you are already using to sample its own authenticated model route in fresh sessions, then compare the answer distributions with public references. No key extraction and no environment-model guessing.
Host-managed authentication
Independent fresh sessions
Paper-inspired JSD
Shortest run prompt01
Original direct-install prompt02
Requires Codex CLI and Node.js 20+. Quick mode uses 20 independent ephemeral generations.
╭────────────────────────────────────────────────────────────────────────────────────╮ │ ◈ SKILL SCAN RESULT verifyllmapi.com DEMO │ ├────────────────────────────────────────────────────────────────────────────────────┤ │ │ │ ◆ CONSISTENT │ │ BEHAVIOR MATCHES CLAIMED PUBLIC REFERENCE │ │ │ ├────────────────────────────────────────────────────────────────────────────────────┤ │ ROUTE BINDING │ │ Claim gpt-5.5 │ │ Child route openai / gpt-5.5 │ │ Model pin gpt-5.5 │ │ Auth host-managed; credential read = NO │ │ Isolation 20/20 fresh ephemeral session IDs │ │ Sampling T=1 host default • reasoning=none • 4 cells × 5 │ ├────────────────────────────────────────────────────────────────────────────────────┤ │ DISTRIBUTION EVIDENCE │ │ ● animal-random|en 5/5 valid mode otter │ │ ● word-random|en 5/5 valid mode serendipity │ │ ● animal-random|ru 5/5 valid mode тигр │ │ ● animal-random|ar 5/5 valid mode أسد │ ├────────────────────────────────────────────────────────────────────────────────────┤ │ PUBLIC REFERENCE COMPARISON Zenodo 10.5281/zenodo.21278557 │ │ Claim ref openai/gpt-5.5 │ │ Mean JSD 0.412684 accept ≤ 0.6439 │ │ Calibration 11.8% EER • same-data, post-hoc cells • OpenRouter cohort │ │ Nearest openai/gpt-5.5 JSD 0.412684 │ ├────────────────────────────────────────────────────────────────────────────────────┤ │ FINGERPRINT fp_session_7c10e8a45d2f │ │ 20/20 valid samples • p50 6840 ms • 248420 Codex tokens │ │ Behavioral consistency, not cryptographic proof or vendor attestation. │ ╰────────────────────────────────────────────────────────────────────────────────────╯
20 fresh sessionsQuick profile, not one self-report
Credential read: noCodex owns the authenticated route
JSD distributionsPaper-inspired behavioral evidence
Open referenceZenodo provenance is included
How it works
One prompt. Three steps. No credential plumbing.
The Agent reads the run contract, downloads the bundled scanner, and delegates every probe to a clean process using its own existing Codex login.
Sample its own route
Each short answer comes from a separate codex exec --ephemeral process. Codex reuses the host login; the scanner never receives it.
Compare distributions
The scanner normalizes the answers, computes Jensen–Shannon divergence, and prints the exact reference, threshold, calibration, and caveats.
Evidence, not vibes
What actually counts as fingerprint evidence
A model name, JSON schema, or ability to follow instructions is not a fingerprint. The evidence comes from repeated empirical answer distributions.
Independent sampling
Low-entropy animal, word, character, and color probes are repeated in fresh sessions rather than answered as one correlated list.
Distribution distance
Per-cell categorical distributions are compared over their joint support with base-2 Jensen–Shannon divergence, then averaged.
Named public reference
The report names the OpenRouter-collected Zenodo reference, selected cell budget, threshold, same-data EER, and transport mismatch risk.
Local-first security
The scanner does not need your Agent credential.
Codex is already authenticated, so the simplest safe design is to let Codex make the calls. The scanner checks login status, starts ephemeral processes, and never opens an auth file or scans key variables.
Uses host-managed Codex login and fresh ephemeral session IDs.
Does not read auth files, cookies, browser storage, process memory, or API-key variables.
Deletes its temporary sample directory after local aggregation.
Keeps explicit endpoint/key mode as an opt-in legacy workflow only.
Download the auditable packageSKILL.md, scanner source, metadata, and license included
FAQ
LLM API verification, without overclaiming
Does Verify LLM API read my Codex login token?
No. The scanner invokes the official Codex CLI, which owns its authentication. It checks only the non-secret output of codex login status and refuses API-key arguments in current-session mode.
Can this prove the exact model with 100% certainty?
No. Behavioral distributions can detect inconsistency and drift, but they are not cryptographic proof or vendor attestation. Missing references and unstable routes return INCONCLUSIVE.
What about temperature 1?
The paper uses temperature 1. Codex's official Responses request does not expose a CLI temperature field and omits it; the Responses API default is 1. The report labels this as a host default, not an explicit override.
How many model generations does a check use?
Quick uses 4 cells × 5 samples = 20 fresh generations. Standard uses 40. Paper mode uses 8 cells × 15 samples = 120.
Can I still test a third-party endpoint directly?
Yes. The original OpenAI-compatible and Anthropic endpoint scanner remains bundled, but it runs only when you explicitly supply the base URL, model, protocol, and credential variable name.
Field notes
A fresh Codex instance tests the production /run path.
The acceptance run starts with an empty workspace and only the short Chinese prompt. It must discover the live protocol, download the production skill, reuse Codex authentication, and preserve an honest verdict.
Read the reproducible acceptance testFresh instance · production artifact · complete terminal evidence
Ask the Agent to test the route it is actually using.
One short prompt, host-managed authentication, and an auditable distribution report.