managing-growthbook

name	managing-growthbook
description	Use when working with Growthbook — growthBook feature flag management, A/B testing, experiment analysis, and data-driven decisions. Covers feature definitions, experiment configuration, metric tracking, visual editor experiments, and SDK connections. Use when managing feature flags, reviewing experiment results, configuring targeting attributes, or analyzing experiment metrics in GrowthBook.
connection_type	growthbook
preload	false

GrowthBook Management Skill

Manage and analyze feature flags, experiments, metrics, and data sources in GrowthBook.

API Conventions

Authentication

All API calls use the Authorization: Bearer $GROWTHBOOK_API_KEY header. Never hardcode tokens.

Base URL

$GROWTHBOOK_URL/api/v1 (cloud: https://api.growthbook.io/api/v1, or self-hosted)

Core Helper Function

#!/bin/bash

GROWTHBOOK_BASE="${GROWTHBOOK_URL:-https://api.growthbook.io}"

gb_api() {
    local method="$1"
    local endpoint="$2"
    local data="${3:-}"

    if [ -n "$data" ]; then
        curl -s -X "$method" \
            -H "Authorization: Bearer $GROWTHBOOK_API_KEY" \
            -H "Content-Type: application/json" \
            "${GROWTHBOOK_BASE}/api/v1${endpoint}" \
            -d "$data"
    else
        curl -s -X "$method" \
            -H "Authorization: Bearer $GROWTHBOOK_API_KEY" \
            -H "Content-Type: application/json" \
            "${GROWTHBOOK_BASE}/api/v1${endpoint}"
    fi
}

Output Rules

• TOKEN EFFICIENCY: Extract only needed fields with jq
• Target <=50 lines per script output
• Never dump full API responses

Discovery Phase

List Features

#!/bin/bash
echo "=== Features ==="
gb_api GET "/features?limit=25" \
    | jq -r '.features[] | "\(.valueType)\t\(.id)\t\(.defaultValue)\t\(.environments | keys | join(","))"' \
    | column -t

echo ""
echo "=== Feature Summary ==="
gb_api GET "/features?limit=100" \
    | jq '{total: (.features | length), by_type: (.features | group_by(.valueType) | map({(.[0].valueType): length}) | add), archived: ([.features[] | select(.archived)] | length)}'

List Experiments

#!/bin/bash
echo "=== Experiments ==="
gb_api GET "/experiments?limit=20" \
    | jq -r '.experiments[] | "\(.status)\t\(.name[0:40])\t\(.variations | length) variations\t\(.phases | length) phases"' \
    | column -t

echo ""
echo "=== Experiment Status Summary ==="
gb_api GET "/experiments?limit=50" \
    | jq -r '.experiments[] | .status' | sort | uniq -c | sort -rn

Analysis Phase

Experiment Details

#!/bin/bash
EXPERIMENT_ID="${1:?Experiment ID required}"

echo "=== Experiment Details ==="
gb_api GET "/experiments/${EXPERIMENT_ID}" \
    | jq '.experiment | {id, name, status, hypothesis: .hypothesis[0:100], variations: [.variations[].name], metrics: .metrics, guardrailMetrics: .guardrailMetrics}'

echo ""
echo "=== Results ==="
gb_api GET "/experiments/${EXPERIMENT_ID}/results" \
    | jq -r '.result | .variations[] | "\(.variationId)\t\(.analyses[0].stats.mean // "pending")"' | column -t 2>/dev/null || echo "Results not yet available"

Metrics and Data Sources

#!/bin/bash
echo "=== Metrics ==="
gb_api GET "/metrics?limit=20" \
    | jq -r '.metrics[] | "\(.type)\t\(.name[0:30])\t\(.datasource[0:20])"' | column -t

echo ""
echo "=== Data Sources ==="
gb_api GET "/data-sources" \
    | jq -r '.dataSources[] | "\(.type)\t\(.name)\t\(.settings.queries | length // 0) queries"' | column -t

echo ""
echo "=== SDK Connections ==="
gb_api GET "/sdk-connections" \
    | jq -r '.connections[] | "\(.name)\t\(.language)\t\(.environment)"' | column -t | head -10

Output Format

• Use tab-separated columns with column -t
• Limit lists to 15-25 items
• Show summaries before details

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

Common Pitfalls

• Self-hosted vs cloud: Base URL varies -- always use $GROWTHBOOK_URL env variable
• Value types: boolean, string, number, json for feature values
• Experiment statuses: draft, running, stopped
• Statistical engine: GrowthBook uses Bayesian or Frequentist analysis -- check organization settings
• Metrics vs guardrails: Guardrail metrics are monitored but not optimized for
• Environments: Features have per-environment rules and enabled state