Evidence row 1747589

What was observed

{
  "count": 4,
  "tools": [
    {
      "description": "Ask the Fittest swarm whether a machine-learned interatomic potential can be trusted on your molecular structures, or whether they need DFT. Three model agents (MACE-MPA-0, AIMNet2, Orb-v3) predict forces; their disagreement sigma_F and the Calibrator's DFT-label ledger give the verdict.\n\nInput `structures`: XYZ text (single or multi-frame; bare 'El x y z' atoms on separate lines or separated by commas/semicolons also work) and/or a list of objects {\"Z\": [8,1,1] or \"symbols\": [\"O\",\"H\",\"H\"], \"R\": [[x,y,z], ...]} in Angstrom. Exact element symbols only. Limits: <= 20 structures per call, 2..100 atoms each, isolated molecules (no periodic cells). Elements the full committee supports: H B C N O F Si P S Cl As Se Br I.\n\nReturns per structure: verdict (trust / send_to_dft) with verdict_reasons, trust_score (Beta posterior P(correct), never exactly 1) with its 90% interval and labels_behind_score, the recommended model or blend, served and reference energy (eV; compare reference_energy_eV across a batch), max/mean force (eV/A), sigma_F vs tau, per-model summary, domain_warning outside C/H/N/O or Transition1x sizes. A bad structure gets its own error entry; the rest are still scored. Rule: trust only if: every model agent supports all elements, the forces do not all vanish, sigma_F < tau, and trust_score >= 0.5. Otherwise send_to_dft, with verdict_reasons. tau is a fixed escalation threshold: the 70th percentile of sigma_F on the round-0 pool. The Calibrator's threshold rule never moved it during the run (see results/RESULTS.md).\n\nLogging: every call to this server is logged (UTC time, tool name, an anonymous client id = salted hash of client name + IP), structure and atom counts, verdict counts, latency and ok/error category. Raw IPs are not stored by the app or written to the access log; since 2026-10-31 19:11 UTC Caddy's error/unmapped-host logs drop them too (before that, some Caddy error and bare-IP entries in the system journal contained client IPs, kept until journal rotation). Structures, coordinates, element lists and names are NOT stored. Aggregate counts are public at /stats.",
      "inputSchema": {
        "properties": {
          "include_forces": {
            "default": false,
            "title": "Include Forces",
            "type": "boolean"
          },
          "structures": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "items": {
                  "anyOf": [
                    {
                      "type": "string"
                    },
                    {
                      "additionalProperties": true,
                      "type": "object"
                    }
                  ]
                },
                "type": "array"
              },
              {
                "additionalProperties": true,
                "type": "object"
              }
            ],
            "title": "Structures"
          }
        },
        "required": [
          "structures"
        ],
        "title": "check_structuresArguments",
        "type": "object"
      },
      "name": "check_structures"
    },
    {
      "description": "Explain why Fittest trusted or escalated ONE structure (XYZ text or {Z|symbols, R}): the verdict rule applied, sigma_F vs tau, the Calibrator bucket with its label count, the posterior P(correct) of every candidate in it, which model agents disagree most, and the most uncertain atoms.\n\nLogging: every call to this server is logged (UTC time, tool name, an anonymous client id = salted hash of client name + IP), structure and atom counts, verdict counts, latency and ok/error category. Raw IPs are not stored by the app or written to the access log; since 2026-10-31 19:11 UTC Caddy's error/unmapped-host logs drop them too (before that, some Caddy error and bare-IP entries in the system journal contained client IPs, kept until journal rotation). Structures, coordinates, element lists and names are NOT stored. Aggregate counts are public at /stats.",
      "inputSchema": {
        "properties": {
          "structure": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "items": {
                  "anyOf": [
                    {
                      "type": "string"
                    },
                    {
                      "additionalProperties": true,
                      "type": "object"
                    }
                  ]
                },
                "type": "array"
              },
              {
                "additionalProperties": true,
                "type": "object"
              }
            ],
            "title": "Structure"
          }
        },
        "required": [
          "structure"
        ],
        "title": "explainArguments",
        "type": "object"
      },
      "name": "explain"
    },
    {
      "description": "Show the Fittest population: model agents (answerers) and the Skeptic (critic) with reputations from the ledger, the calibration domain, tau and the verdict rule, which models are live, and the evolution family trees if published.\n\nLogging: every call to this server is logged (UTC time, tool name, an anonymous client id = salted hash of client name + IP), structure and atom counts, verdict counts, latency and ok/error category. Raw IPs are not stored by the app or written to the access log; since 2026-10-31 19:11 UTC Caddy's error/unmapped-host logs drop them too (before that, some Caddy error and bare-IP entries in the system journal contained client IPs, kept until journal rotation). Structures, coordinates, element lists and names are NOT stored. Aggregate counts are public at /stats.",
      "inputSchema": {
        "properties": {},
        "title": "swarm_statusArguments",
        "type": "object"
      },
      "name": "swarm_status"
    },
    {
      "description": "Aggregate usage of this server: distinct users (anonymous clients that got at least one structure verdict; internal test clients excluded), total calls, calls per tool, first/last call time. No per-user data.\n\nLogging: every call to this server is logged (UTC time, tool name, an anonymous client id = salted hash of client name + IP), structure and atom counts, verdict counts, latency and ok/error category. Raw IPs are not stored by the app or written to the access log; since 2026-10-31 19:11 UTC Caddy's error/unmapped-host logs drop them too (before that, some Caddy error and bare-IP entries in the system journal contained client IPs, kept until journal rotation). Structures, coordinates, element lists and names are NOT stored. Aggregate counts are public at /stats.",
      "inputSchema": {
        "properties": {},
        "title": "usage_statsArguments",
        "type": "object"
      },
      "name": "usage_stats"
    }
  ]
}

What the leaf hashes

{"body":"3250c6aac88ca48a8f9c2ebc4dc9a6fcc40e11352d643ce3a535796b0a1c2da9","id":1747589,"kind":"mcp.tools","method":"probe.mcp/1","observed_us":1793516638664858,"prev":null,"subject":"7cab0d9c-03dd-4336-9cad-150c27c30fe8","subject_kind":"endpoint"}

Proof

Its day has not closed yet. A day closes ten minutes after midnight UTC; the proof appears here then.