managing-hypertrace

name	managing-hypertrace
description	Use when working with Hypertrace — hypertrace distributed tracing and observability platform for service dependency mapping, trace analysis, API monitoring, and performance analysis. Covers service discovery, endpoint performance, trace querying, span analysis, and API-level observability. Use when investigating distributed traces, analyzing API performance, mapping service dependencies, or reviewing span-level details.
connection_type	hypertrace
preload	false

Hypertrace Monitoring Skill

Query, analyze, and manage Hypertrace observability data using the Hypertrace GraphQL API.

API Overview

Hypertrace uses a GraphQL API at https://<HYPERTRACE_HOST>/graphql.

Core Helper Function

#!/bin/bash

ht_gql() {
    local query="$1"
    curl -s -X POST "${HYPERTRACE_URL}/graphql" \
        -H "Content-Type: application/json" \
        ${HYPERTRACE_AUTH:+-H "Authorization: Bearer $HYPERTRACE_AUTH"} \
        -d "{\"query\": $(echo "$query" | jq -Rs .)}"
}

ht_time_range() {
    local hours="${1:-1}"
    local end=$(($(date +%s) * 1000))
    local start=$(( ($(date +%s) - ${hours} * 3600) * 1000 ))
    echo "startTime: \"${start}\", endTime: \"${end}\""
}

MANDATORY: Discovery-First Pattern

Always discover services, APIs, and backends before querying.

Phase 1: Discovery

#!/bin/bash
TIME_RANGE=$(ht_time_range 1)

echo "=== Services ==="
ht_gql "{
    explore(scope: API_TRACE, limit: 100, groupBy: [{key: \"serviceName\"}],
        selections: [{key: \"serviceName\"}, {key: \"numCalls\", aggregation: SUM}],
        timeRange: {${TIME_RANGE}}) {
        results { serviceName numCalls }
    }
}" | jq -r '.data.explore.results[] | "\(.serviceName)\tcalls:\(.numCalls)"' | sort -t$'\t' -k2 -rn | head -20

echo ""
echo "=== APIs / Endpoints ==="
ht_gql "{
    explore(scope: API, limit: 30, groupBy: [{key: \"apiName\"}],
        selections: [{key: \"apiName\"}, {key: \"serviceName\"}, {key: \"numCalls\", aggregation: SUM}],
        timeRange: {${TIME_RANGE}}) {
        results { apiName serviceName numCalls }
    }
}" | jq -r '.data.explore.results[] | "\(.serviceName)\t\(.apiName)\tcalls:\(.numCalls)"' | head -20

echo ""
echo "=== Backends ==="
ht_gql "{
    explore(scope: BACKEND, limit: 20, groupBy: [{key: \"backendName\"}],
        selections: [{key: \"backendName\"}, {key: \"backendType\"}, {key: \"numCalls\", aggregation: SUM}],
        timeRange: {${TIME_RANGE}}) {
        results { backendName backendType numCalls }
    }
}" | jq -r '.data.explore.results[] | "\(.backendName)\t\(.backendType)\tcalls:\(.numCalls)"' | head -15

Phase 2: Analysis

#!/bin/bash
TIME_RANGE=$(ht_time_range 1)

echo "=== Service Performance (P99 Latency) ==="
ht_gql "{
    explore(scope: API_TRACE, limit: 15, groupBy: [{key: \"serviceName\"}],
        selections: [{key: \"serviceName\"}, {key: \"duration\", aggregation: P99}, {key: \"numCalls\", aggregation: SUM}, {key: \"errorCount\", aggregation: SUM}],
        timeRange: {${TIME_RANGE}}, orderBy: [{key: \"duration\", aggregation: P99, order: DESC}]) {
        results { serviceName duration numCalls errorCount }
    }
}" | jq -r '.data.explore.results[] | "\(.serviceName)\tp99:\(.duration / 1000 | . * 10 | round / 10)ms\tcalls:\(.numCalls)\terrors:\(.errorCount)"' | head -15

echo ""
echo "=== Slowest Endpoints ==="
ht_gql "{
    explore(scope: API, limit: 15, groupBy: [{key: \"apiName\"}, {key: \"serviceName\"}],
        selections: [{key: \"apiName\"}, {key: \"serviceName\"}, {key: \"duration\", aggregation: AVG}, {key: \"numCalls\", aggregation: SUM}],
        timeRange: {${TIME_RANGE}}, orderBy: [{key: \"duration\", aggregation: AVG, order: DESC}]) {
        results { apiName serviceName duration numCalls }
    }
}" | jq -r '.data.explore.results[] | "\(.serviceName)\t\(.apiName[0:40])\tavg:\(.duration / 1000 | . * 10 | round / 10)ms\tcalls:\(.numCalls)"' | head -15

echo ""
echo "=== Error Endpoints ==="
ht_gql "{
    explore(scope: API, limit: 15, groupBy: [{key: \"apiName\"}, {key: \"serviceName\"}],
        selections: [{key: \"apiName\"}, {key: \"serviceName\"}, {key: \"errorCount\", aggregation: SUM}, {key: \"numCalls\", aggregation: SUM}],
        timeRange: {${TIME_RANGE}}, filterBy: [{key: \"errorCount\", operator: GT, value: \"0\"}],
        orderBy: [{key: \"errorCount\", aggregation: SUM, order: DESC}]) {
        results { apiName serviceName errorCount numCalls }
    }
}" | jq -r '.data.explore.results[] | "\(.serviceName)\t\(.apiName[0:40])\terrors:\(.errorCount)\tcalls:\(.numCalls)"' | head -15

Output Rules

• TOKEN EFFICIENCY: Target ≤50 lines — use limit and orderBy for top-N results
• Duration values are in microseconds — divide by 1000 for ms
• Scopes: API_TRACE (service-level), API (endpoint-level), BACKEND (dependency-level)
• Use filterBy for server-side filtering and groupBy for aggregation

Output Format

Present results as a structured report:

Managing Hypertrace Report
══════════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]

Target ≤50 lines of output. Use tables for multi-resource comparisons.

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative