managing-statsig

name	managing-statsig
description	Use when working with Statsig — statsig feature gate management, dynamic configs, experiment analysis, and metric tracking. Covers gate configuration, rule-based targeting, holdout groups, A/B test results, pulse metrics, and layer management. Use when managing feature gates, reviewing experiment results, analyzing metric lifts, or configuring dynamic configs in Statsig.
connection_type	statsig
preload	false

Statsig Management Skill

Manage and analyze feature gates, experiments, dynamic configs, and metrics in Statsig.

API Conventions

Authentication

All API calls use the statsig-api-key: $STATSIG_CONSOLE_API_KEY header (Console API key). Never hardcode tokens.

Base URL

https://statsigapi.net/console/v1

Core Helper Function

#!/bin/bash

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

    if [ -n "$data" ]; then
        curl -s -X "$method" \
            -H "statsig-api-key: $STATSIG_CONSOLE_API_KEY" \
            -H "Content-Type: application/json" \
            "https://statsigapi.net/console/v1${endpoint}" \
            -d "$data"
    else
        curl -s -X "$method" \
            -H "statsig-api-key: $STATSIG_CONSOLE_API_KEY" \
            -H "Content-Type: application/json" \
            "https://statsigapi.net/console/v1${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 Gates and Configs

#!/bin/bash
echo "=== Feature Gates ==="
statsig_api GET "/gates" \
    | jq -r '.data[] | "\(.isEnabled)\t\(.name)\t\(.rules | length) rules"' | column -t | head -25

echo ""
echo "=== Dynamic Configs ==="
statsig_api GET "/dynamic_configs" \
    | jq -r '.data[] | "\(.isEnabled)\t\(.name)"' | column -t | head -15

List Experiments

#!/bin/bash
echo "=== Experiments ==="
statsig_api GET "/experiments" \
    | jq -r '.data[] | "\(.status)\t\(.name[0:40])\t\(.groups | length) groups"' | column -t | head -20

echo ""
echo "=== Experiment Status Summary ==="
statsig_api GET "/experiments" \
    | jq -r '.data[] | .status' | sort | uniq -c | sort -rn

Analysis Phase

Gate Details

#!/bin/bash
GATE_NAME="${1:?Gate name required}"

echo "=== Gate Details ==="
statsig_api GET "/gates/${GATE_NAME}" \
    | jq '.data | {name, isEnabled, description: .description[0:100], rules: [.rules[] | {name, passPercentage, conditions: (.conditions | length)}]}'

Experiment Results

#!/bin/bash
EXPERIMENT_NAME="${1:?Experiment name required}"

echo "=== Experiment Details ==="
statsig_api GET "/experiments/${EXPERIMENT_NAME}" \
    | jq '.data | {name, status, hypothesis: .hypothesis[0:100], groups: [.groups[].name], primaryMetrics: .primaryMetrics}'

echo ""
echo "=== Pulse Results ==="
statsig_api GET "/experiments/${EXPERIMENT_NAME}/pulse" \
    | jq -r '.data.results[0:10][] | "\(.metric_name[0:30])\t\(.test_group)\tlifted:\(.lift // "pending")\tp:\(.p_value // "pending")"' \
    | column -t 2>/dev/null || echo "Results not yet available"

Layers

#!/bin/bash
echo "=== Layers ==="
statsig_api GET "/layers" \
    | jq -r '.data[] | "\(.name)\t\(.experiments | length) experiments"' | column -t | head -15

echo ""
echo "=== Holdout Groups ==="
statsig_api GET "/holdouts" \
    | jq -r '.data[] | "\(.isEnabled)\t\(.name)\t\(.holdoutPercentage)%"' | column -t

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

• Console vs server API: Console API (statsig-api-key header) for management; Server API for evaluation
• Gate rules: Rules are evaluated top-to-bottom, first matching rule wins
• Layers: Layers allow mutually exclusive experiments -- users assigned to one experiment per layer
• Holdouts: Holdout groups exclude users from all experiments for measurement
• Pulse: Experiment results are in the pulse endpoint -- may take time to compute
• Rate limits: Console API has rate limits -- check response headers