Detection results

Model: · mode · relay

Total score

Each item is pass / fail / error; the total is a weighted result and may be in-between.

0 / 100

Model authentication Verification completed

Examine the model's self-reported identity, brand residue, and common signs of camouflage.

Not wired —

Think supportive Verification completed

Verify whether server-side exclusive fields such as Claude thinking / signature are transparently transmitted.

Not wired —

Accurate command reproduction (anti-casing) Verification completed

The model is asked to reproduce a fixed mark verbatim, checking how deterministic the instructions are followed.

Not wired —

System conflict testing Verification completed

Compare user only to explicit system+user, checking whether fixed words are overridden by hidden directives.

Not wired —

Extreme reasoning ability (number theory counting) Verification completed

Use canned reasoning questions to check whether verifiable answers are correct.

Not wired —

Channel detection Verification completed

Identifies the hosting channel from the response ID prefix.

Not wired —

Claude Tokenizer verification Verification completed

Count input_tokens with fixed hints compared to calibrated model baseline.

Not wired —

Minimalist Input Token Audit Verification completed

Use minimalist requests to observe whether input_tokens is abnormally high and identify implicit System injection.

Not wired —

behavioral style consistency Verification completed

Compare Claude series characteristics through behavioral preferences and answer patterns.

Not wired —

knowledge accuracy Verification completed

Use Anthropic / Claude general knowledge questions to cross-validate whether the backend is as claimed as Claude.

Not wired —

Model declaration consistency Verification completed

Comparing request model, response model and identity stability in multiple rounds of returns.

Not wired —

Message structure specification Verification completed

Check message IDs, SSE event sequences, and content blocks for compliance with Anthropic specifications.

Not wired —

Tool Use/Structured Output Verification completed

Check whether the tool_use block, toolu_ ID, function name and parameter schema are standardized.

Not wired —

streaming consistency Verification completed

Compare stream and non-stream text, token, and end reason for consistency.

Not wired —

Token usage consistency Verification completed

Check that the input / output / cache token fields are complete and authentic.

Not wired —

Official comparison of the same title Verification completed

The same question compares input consumption and response behavior between channels and official (or baseline).

Not wired —

Hide injection detection Verification completed

Randomized variants cross-check for abnormal signals such as nonce leaks, fixed replies, and suspicious cache reuse.

Not wired —

Security rejection and policy stability Verification completed

Check if the security policy is stable using denial-of-answer scenarios and rewriting/encoding/progressive variants.

Not wired —

long context authenticity Verification completed

Use needle-in-haystack to verify that long context windows are honored.

Not wired —

Multimodal/PDF recognition Verification completed

Submit a PDF probe to confirm that the document understanding link has not been stripped by the transfer station.

Not wired —

First token

—

Total time

0ms

Throughput (T/S)

—

Input tokens

Output tokens

What does each Claude check cover?

Model authentication: Examine the model's self-reported identity, brand residue, and common signs of camouflage.
Think supportive ⭐: Verify whether server-side exclusive fields such as Claude thinking / signature are transparently transmitted.
Accurate command reproduction (anti-casing): The model is asked to reproduce a fixed mark verbatim, checking how deterministic the instructions are followed.
System conflict testing: Compare user only to explicit system+user, checking whether fixed words are overridden by hidden directives.
Minimalist Input Token Audit: Use minimalist requests to observe whether input_tokens is abnormally high and identify implicit System injection.
Extreme reasoning ability (number theory counting): Use canned reasoning questions to check whether verifiable answers are correct.
Channel detection: Identifies the hosting channel from the response ID prefix.
Claude Tokenizer verification: Count input_tokens with fixed hints compared to calibrated model baseline.
behavioral style consistency: Compare Claude series characteristics through behavioral preferences and answer patterns.
knowledge accuracy: Use Anthropic / Claude general knowledge questions to cross-validate whether the backend is as claimed as Claude.
Model declaration consistency: Comparing request model, response model and identity stability in multiple rounds of returns.
Message structure specification: Check message IDs, SSE event sequences, and content blocks for compliance with Anthropic specifications.
Tool Use/Structured Output: Check whether the tool_use block, toolu_ ID, function name and parameter schema are standardized.
streaming consistency: Compare stream and non-stream text, token, and end reason for consistency.
Token usage consistency: Check that the input / output / cache token fields are complete and authentic.
Official comparison of the same title: The same question compares input consumption and response behavior between channels and official (or baseline).
long context authenticity: Use needle-in-haystack to verify that long context windows are honored.
Multimodal/PDF recognition: Submit a PDF probe to confirm that the document understanding link has not been stripped by the transfer station.
Hide injection detection: Randomized variants cross-check for abnormal signals such as nonce leaks, fixed replies, and suspicious cache reuse.
Security rejection and policy stability: Check if the security policy is stable using denial-of-answer scenarios and rewriting/encoding/progressive variants.