Leaderboard / sweetai.tiamatcrypto.com / endpoint
GET https://sweetai.tiamatcrypto.com/api/v1/trust-forensics
Scores measure reliability and honesty, not output quality.
- The quoted payment wallet does not match the catalog listing.
- Listed only 1 day ago; new listings earn trust over 30 days.
- Category
- Signal by llm, confidence 60%
- Latest quote
- $0.01 on Base to 0xB96A…3DFb
- Uptime
- 100% / 7d, 100% / 30d
- HTTPS
- yes
- First listed
- 2026-11-16
- Last probe
- 3h ago
- Next probe
- 2026-11-17 16:55 UTC
Score components
Catalog listings
| Source | Scheme | Network | Asset | Listed price | Pay to | Flow | Current | Last seen |
|---|---|---|---|---|---|---|---|---|
| cdp | exact | Polygon | 0x3c49…3359 | 10000 | 0x2D03…B6ca | default | yes | 41m ago |
| cdp | exact | Arbitrum | 0xaf88…5831 | 10000 | 0x2D03…B6ca | default | yes | 41m ago |
| cdp | exact | Solana | EPjFWd…Dt1v | 10000 | EiaX7k…cDeH | default | yes | 41m ago |
| payai | exact | Solana | EPjFWd…Dt1v | 5000 | EiaX7k…cDeH | default | yes | 9m ago |
| cdp | exact | Base | 0x8335…2913 | $0.01 | 0x2D03…B6ca | default | yes | 41m ago |
Paid probes
No paid probes yet. Paid probing is off by default and runs weekly per host under a strict spend cap; each paid probe links to its on-chain transaction here.
Delivery checks
Inside each paid probe we ask for something whose correct answer we know, then compare. A pass means the seller delivered what it was paid for; a fail means it charged and returned something wrong; inconclusive means we could not judge (no reference, unknown input shape, or a non-2xx response). Every row carries the request, the response hash and the reference so the verdict can be reproduced. Delivery results do not affect scores yet.
No delivery checks yet. They run inside the weekly paid probe.
Classification history
Every classification is kept; the newest row is the current category. Rules match the listing text; the model is consulted only when the rules are unsure.
| When | Category | Confidence | Method | Version | Evidence |
|---|---|---|---|---|---|
| 13h ago | Signal | 60% | llm | v1 | claude-haiku-5-5: The endpoint returns a forensic verdict on whether an AI agent's reputation is genuine, flagging Sybil patterns, ghost reviewers and bought feedback. That is a risk verdict rather than raw chain data, so signal fits best, though it reads on-chain registry data, so chain_read is a plausible alternative. |
| 17h ago | Other | 40% | rules | v1 | rule none matched "" in none |
Probe history
| When | Kind | Status | Latency | x402 | Valid | Quoted | Pay to | Flags |
|---|---|---|---|---|---|---|---|---|
| 3h ago | free | 402 | 1218 ms | v2 | yes | $0.01 | 0xB96A…3DFb | payto_mismatch |
| 10h ago | free | 402 | 251 ms | v2 | yes | $0.01 | 0xB96A…3DFb | payto_mismatch |
| 16h ago | free | 402 | 682 ms | v2 | yes | $0.01 | 0xB96A…3DFb | payto_mismatch |
| 22h ago | free | 402 | 1257 ms | v2 | yes | $0.01 | 0xB96A…3DFb | payto_mismatch |
| 28h ago | free | 402 | 229 ms | v2 | yes | $0.01 | 0xB96A…3DFb | payto_mismatch |
| 35h ago | free | 402 | 633 ms | v2 | yes | $0.01 | 0xB96A…3DFb | payto_mismatch |