Skip to content

Examples

Given a pattern (or a pasted log line that resolves to one), pull recent live events from the log analyzer and group them by exact template. Returns the top 3 buckets by event count with per-bucket template label, recall, and extracted slot values per match. Bounded to a 24-hour window in the live analyzer. When a pattern is under an offload action, the SIEM never received those events; to sample the held-back cohort instead, see Retriever query.

Typically called by Investigate when a metric tier surfaces a pattern that needs event evidence to form a hypothesis. Not usually invoked directly.

Example

You

show examples of Payment_Gateway_Timeout from the last hour

Log10x

From Datadog, 1h, 12 events, 3 template buckets:

bucket template events
1 Payment_Gateway_Timeout: connect timeout after {ms}ms on host {host} 7
2 Payment_Gateway_Timeout: read timeout, retried {n} times 3
3 Payment_Gateway_Timeout: SSL handshake failed: {error} 2

Bucket 1 sample: ms=4823, host=payments-3.acme.net · ms=5102, host=payments-7.acme.net · …

More to ask

  • "examples of this pasted line, scope to cart-svc, error level only"
  • "slow examples, show 50 per bucket"

Prerequisites

Log analyzer credentials in the env for one of: Splunk, Datadog, Elasticsearch, CloudWatch. When multiple are configured, pass vendor= explicitly.

Schema and samples

Input example

Real call against the demo env (captured by scripts/capture-tool-envelopes.mjs).

{
  "pattern": "open_telemetry_opensearchexporter_clientLogger_LogRoundTrip_open_telemetry_opensearchexporter_v_go_github_opensearch_project",
  "limit": 3,
  "timeRange": "1h",
  "view": "summary"
}
Input schema

Agent-facing JSON Schema (the canonical shape the MCP server publishes via tools/list):

{
  "type": "object",
  "properties": {
    "pattern": {
      "type": "string",
      "description": "Pattern name (Symbol Message, e.g. `Payment_Gateway_Timeout`) or a pasted raw log line. Pasted lines resolve to the matching pattern via the same templater path as log10x_resolve_batch. Required."
    },
    "vendor": {
      "type": "string",
      "enum": [
        "splunk",
        "datadog",
        "elasticsearch",
        "cloudwatch"
      ],
      "description": "Log analyzer to search. Auto-detected when exactly one of the supported vendors has credentials in the env; pass explicitly when multiple are configured."
    },
    "service": {
      "type": "string",
      "description": "Optional service-name scope. Translated to the vendor-specific service filter."
    },
    "severity": {
      "type": "string",
      "description": "Optional severity scope (e.g., `ERROR`, `WARN`)."
    },
    "timeRange": {
      "type": "string",
      "enum": [
        "15m",
        "1h",
        "6h",
        "24h",
        "1d",
        "7d",
        "30d"
      ],
      "default": "1h",
      "description": "Window for the live SIEM probe. Capped at 24h. To sample a pattern's offloaded cohort (events the Receiver routed to the overflow bucket, which the SIEM never received), use log10x_retriever_query. '1d' is a legacy alias for '24h'."
    },
    "limit": {
      "type": "number",
      "minimum": 1,
      "maximum": 50,
      "default": 10,
      "description": "Maximum number of sample events per templateHash bucket. 1-50, default 10."
    },
    "scope": {
      "type": "string",
      "description": "Vendor-specific scope (Splunk index, Datadog index, ES index pattern, CloudWatch log group). Defaults to a sensible per-vendor value when omitted."
    },
    "environment": {
      "type": "string",
      "description": "Environment nickname."
    }
  },
  "required": [
    "pattern"
  ],
  "additionalProperties": false
}

Source: src/tools/pattern-examples.ts.

Output example

Real envelope from the demo env. view: "summary" returns the full StructuredOutput with typed data. Long arrays + base64 PNG bodies trimmed for readability; the real call returns them in full.

Headline (the 1-line agent-facing answer):

open\_telemetry\_opensearchexporter\_clientLogger\_LogRoundTrip\_open\_telemetry\_opensearchexporter\_v\_go\_github\_opensearch\_project (cloudwatch, 1h): 102 events pulled, 102 retained across 1 templates via tenx_hash-exact

{
  "schema_version": "1.0",
  "schema_epoch": "2026-05-25",
  "tool": "log10x_pattern_examples",
  "generated_at": "2026-05-26T15:37:29.611Z",
  "view": "summary",
  "summary": {
    "headline": "`open_telemetry_opensearchexporter_clientLogger_LogRoundTrip_open_telemetry_opensearchexporter_v_go_github_opensearch_project` (cloudwatch, 1h): 102 events pulled, 102 retained across 1 templates via tenx_hash-exact"
  },
  "data": {
    "pattern": "open_telemetry_opensearchexporter_clientLogger_LogRoundTrip_open_telemetry_opensearchexporter_v_go_github_opensearch_project",
    "vendor": "cloudwatch",
    "window": "1h",
    "probe_path": "tenx_hash-exact",
    "events_pulled": 102,
    "distinct_templates": 1,
    "retained_events": 102,
    "retained_templates": 1,
    "dropped_jaccard_events": 0,
    "multi_line_detected": false,
    "buckets": [
      {
        "rank": 1,
        "template_hash": "2qiwE{k$rY",
        "tenx_hash": "4Fwju8mGbu4",
        "event_count": 102,
        "jaccard": 1,
        "severity": "ERROR",
        "service": "tenx-fluentd",
        "sample_event": "2026-05-26T14:37:40.275Z\terror\topensearchexporter@v0.142.0/logger.go:36\tRequest failed.\t{\"resource\": {\"service.instance.id\": \"1aab212a-1c9e-423a-9b98-cc7cd26c17ae\", \"service.name\": \"otelcol-contrib\", ",
        "slot_distribution": [
          {
            "slot": "timestamp",
            "distinct_count": 20,
            "is_constant": false,
            "sample_values": [
              "1779806260275",
              "1779806260481",
              "1779806260684"
            ]
          },
          {
            "slot": "slot_9",
            "distinct_count": 20,
            "is_constant": false,
            "sample_values": [
              "002150637",
              "005096784",
              "00205715"
            ]
          },
          {
            "slot": "slot_1",
            "distinct_count": 1,
            "is_constant": true,
            "sample_values": [
              "142"
            ]
          },
          "... 3 more elided"
        ]
      }
    ],
    "probe_notes": [
      "scope auto-discovered: 4 group(s) under \"/log10x*\""
    ]
  },
  "actions": [],
  "truncated": true,
  "warnings": []
}
Output schema

The data block inside the StructuredOutput envelope:

interface ToolData {
  pattern: string;
  vendor: string;
  window: string;
  probe_path: string;
  events_pulled: number;
  distinct_templates: number;
  retained_events: number;
  retained_templates: number;
  dropped_jaccard_events: number;
  multi_line_detected: boolean;
  buckets: Array<{
    rank: number;
    template_hash: string;
    tenx_hash: string;
    event_count: number;
    jaccard: number;
    severity: string;
    service: string;
    sample_event: string;
    slot_distribution: Array<{
    slot: string;
    distinct_count: number;
    is_constant: boolean;
    sample_values: string[];
  }>;
  }>;
  probe_notes: string[];
}

Envelope-level fields the agent should also read: summary.headline (1-line answer), actions[] (next-call chain hints as {tool, args, reason}), truncated: boolean, images[] (PNG attachments where applicable), schema_epoch (engine-ID stability boundary).