Research & Analysis Workflows

Build and run a GrowthBook Product Analytics chart via the REST API — visualize a metric over time, aggregate a fact table, or chart a raw warehouse table, then return the numbers plus a deep link to the chart. Use when the user asks "show me signups by country", "chart daily active users", "how many orders last week", "plot revenue over time", "break that down by plan", or any "show me / chart / plot / how many" question about product data. For discovering what metrics and fact tables exist first, use metric-search. For experiment results, use experiment-analyze — this skill is for general analytics, not A/B test readouts.

Fetch results for a GrowthBook experiment, refresh the snapshot only when the cached data is over 24 hours old, then interpret. Use when the user asks "what are the results of X", "analyze this experiment", "is X winning", "did the test work", "show me the results", or "dig into the dimensions". Reads only — does not stop or modify the experiment. For stopping after you've seen results, use experiment-stop.

Propose new experiment ideas grounded in the team's past stopped experiments via the GrowthBook REST API. Use when the user asks "what should we test next", "give me experiment ideas", "brainstorm A/B tests", "what's worth testing", or "ideas for experiments". Proposes only — does not create experiments. For designing a specific test, use experiment-design. For reading results of one, use experiment-analyze.

Launch a GrowthBook A/B test end-to-end via the REST API — create the experiment, prep the feature flag, wire the experiment-ref rule, and start the experiment. Use when the user says "launch this experiment", "create the experiment", "wire up the A/B test", "kick off the test", "set up X as an experiment in GrowthBook", "start the experiment for flag Y", or "I already have a flag and want to run an experiment on it". Works for both experiment-first (creates the flag) and flag-first (detects the existing flag via the reuse path and wires the experiment to it). For designing the spec first, use experiment-design. For stopping a running experiment, use experiment-stop. For interpreting results, use experiment-analyze.

Stop a running GrowthBook experiment via the REST API, optionally declaring a winning variation and rolling it out to 100% of eligible traffic. Use when the user says "stop this experiment", "end the A/B test", "declare a winner for X", "ship the winning variation", "roll back the test", or "we're done with this experiment". For interpreting results before deciding, use experiment-analyze first.

Archive or delete a stale GrowthBook feature flag, walking the user through inlining the flag's effective value at call sites in the codebase before removal. Use when the user says "delete this flag", "remove this stale flag", "clean up flag X", "archive this flag", "we don't need this flag anymore", or "get rid of this flag and its experiment-ref rule". For finding stale flags first, use flag-search. For editing rules instead of removing the flag, use flag-targeting. For stopping an experiment that uses the flag, use experiment-stop.

Set the default value of a GrowthBook feature flag — the value served when no rules match. Use when the user says "change the default value of flag X", "what does this flag return when no rules apply", "set the fallback to false", "update the default to the new config", or "change what users get by default". Goes through a draft revision and requires publishing. For adding or editing targeting rules, use flag-targeting or flag-rules. For updating administrative metadata, use flag-metadata. For creating a new flag, use flag-create.

Add an experiment-ref rule to a GrowthBook feature flag to run an A/B test through that flag. Use when the user says "run an experiment through this flag", "add an A/B test to flag X", "link experiment Y to this flag", "set up variations on this flag for an experiment", "add an experiment-ref rule", or "wire up this experiment to the flag". For launching a full new experiment end-to-end, use experiment-launch. For editing the targeting conditions or scope of an existing experiment rule, use flag-targeting. For stopping an experiment and cleaning up its rule, use experiment-stop then flag-rules.

Trace the dependency relationships around a GrowthBook feature flag — what it depends on (prerequisites), what depends on it (reverse lookup), which experiments link to it, and any holdout associations. Use when the user asks "what depends on flag X", "what flags does flag X require", "will deleting this flag break anything", "show me the dependency graph", "what experiments are linked to this flag", "find everything that uses flag X", or "is it safe to clean up this flag". Read-only — for making changes to prerequisites, use flag-prerequisites. For cleanup, use flag-cleanup.

Update the administrative metadata of an existing GrowthBook feature flag — description, owner, project, tags, custom fields, or JSON schema. Use when the user says "change the owner of flag X", "add tags to this flag", "move this flag to project Y", "update the description", "set a JSON schema on this flag", "rename the owner", or "tag this flag as payments". Goes through a draft revision and requires publishing to take effect. For changing the flag's default value, use flag-default-value. For creating a new flag, use flag-create. For archiving or deleting a flag, use flag-cleanup.

Set up a monitored progressive rollout ("safe rollout") for a GrowthBook feature flag — combining a ramp schedule with guardrail metric monitoring, automated signals, and optional auto-rollback. Use when the user says "roll this out safely", "monitor the rollout with guardrail metrics", "set up a safe rollout", "I want to ramp this with automatic rollback if metrics regress", "configure monitoring on the ramp", "check the monitoring status of this rollout", "approve the next monitored step", or "roll back because guardrails are failing". For unmonitored ramps (just progressive coverage, no metrics), use flag-ramp directly. For simple on/off time windows, use flag-schedule.

Add, remove, or inspect feature-level prerequisites on a GrowthBook feature flag. Use when the user says "gate flag X on flag Y being enabled", "add a prerequisite", "flag X should only evaluate if flag Y is on", "remove the prerequisite on flag X", "what does this flag depend on", or "this flag should require the new-checkout flag to be true first". Feature-level prerequisites gate the entire flag — when the prerequisite flag is off, this flag returns its default value for every user. The prerequisite must be a boolean flag. For prerequisites scoped to a single rule (not the whole flag), use flag-targeting. For tracing the full dependency graph, use flag-graph.

Publish a GrowthBook feature flag draft revision, resolve merge conflicts, revert to a prior revision, or discard a draft. Use when the user says "publish this draft", "push this live", "go live with revision X", "there's a merge conflict on my flag", "rebase my draft", "fix the merge conflict on flag X", "revert flag X to a previous version", "roll back this flag change", "discard this draft", or "abandon these changes". For requesting or submitting an approval review before publish, use flag-review. For listing and inspecting drafts, use flag-revisions.

Create or manage a multi-step ramp schedule for a GrowthBook feature flag rule — progressively increasing traffic exposure over time with defined intervals between steps. Use when the user says "gradually roll this out", "increase traffic from 5% to 100% over a week", "set up a ramp schedule", "advance to the next ramp step", "pause the rollout", "roll back the ramp", or "set a cutoff date on this rollout". For rollouts that also need guardrail metric monitoring and automatic signals, use flag-monitoring — it builds on this skill with monitoring configuration. For simple on/off time windows, use flag-schedule.

Request or submit an approval review on a GrowthBook feature flag draft revision. Use when the user says "request review for this draft", "approve this change", "reject this draft", "request changes on revision X", "I want to review flag Y's pending draft", "submit my approval", "mark this as needing changes", "who needs to approve this", or "check the review status". For creating and editing drafts, use the relevant flag- write skill. For publishing an approved draft or resolving merge conflicts, use flag-publish. For listing all pending drafts across flags, use flag-revisions.

List, inspect, and manage draft revisions for a GrowthBook feature flag. Use when the user asks "what drafts are open for flag X", "show me pending approvals", "who owns this draft", "discard this draft", "start a new draft", "what revision is my flag on", "list all my pending changes", "what's waiting for review", or any question about the state of in-progress changes to a flag. Every flag write operation — rules, metadata, default value, toggles — works through a draft revision; use this skill to see what's in flight before making changes. For requesting or submitting an approval review, use flag-review. For publishing a draft live or resolving merge conflicts, use flag-publish. For making flag changes, use the relevant flag- write skill.

Entry point for working with rules on a GrowthBook feature flag. Use when the user asks "what rules does flag X have", "show me the rules on this flag", "add a rule", "delete a rule", "reorder the rules", or describes a flag operation without specifying the rule type. Routes to specialized skills for creating and editing specific rule types. Also handles listing, reordering, and deleting rules directly. For specific rule types, use flag-targeting for force/rollout rules, flag-experiment for experiment-ref rules, flag-schedule for timed activation, flag-ramp for progressive rollouts, flag-monitoring for monitored rollouts, flag-prerequisites for feature-level prerequisite gates.

Add a timed activation window to a GrowthBook feature flag rule — automatically enable it at a start time and/or disable it at an end time. Use when the user says "turn this on at 9am", "schedule the flag to go live Friday at noon", "disable the rule after the sale ends", "set an end date on this rule", "run this rule during the promotion window", or "time-gate this rule". Applies to force and rollout rules. For multi-step progressive rollouts with intervals between steps, use flag-ramp. For the broader campaign of rules around this schedule, use flag-targeting first.

Search, list, and audit GrowthBook feature flags. Use when the user asks "what flags do we have", "find flags tagged payments", "list flags in project X", "which flags are stale", "find flags owned by bryce@company.com", "show me all disabled flags in production", "audit our flags", "find flags with no rules", or "what can we clean up". Read-only — for actually removing flags, use flag-cleanup. For inspecting what depends on a specific flag, use flag-graph. For adding or changing rules, use flag-targeting or flag-rules.

Add, edit, or remove force-value or percentage-rollout rules on a GrowthBook feature flag, including conditions, saved groups, and rule-level prerequisites. Use when the user says "add a rule to flag Y", "release this flag to 10% of users", "turn this on for beta testers", "target US users", "add a condition to rule X", "edit the targeting condition on flag Z", "disable this rule", "remove the rule", "only for logged-in users", or "target users matching attribute Y". For environment kill switches (enable/disable the whole flag in an env), use flag-toggle. For experiment rules, use flag-experiment. For ramp schedules, use flag-ramp. For monitored rollouts with guardrail metrics, use flag-monitoring. For feature-level prerequisites, use flag-prerequisites.

Enable or disable a GrowthBook feature flag in a specific environment. Use when the user says "turn on flag X in production", "disable this flag in staging", "kill switch flag X", "turn this off everywhere", "enable the flag", "flip it off in prod", or "I need to disable this flag now". This is the environment-level kill switch — it controls whether the flag is evaluated at all in an environment, independent of its rules. Changes are review-gated and go through the standard draft → publish flow. For adding or editing targeting rules within an environment, use flag-targeting. For toggling a rule on or off without affecting the environment, use flag-rules.

Configure the GrowthBook plugin's API credentials so every other skill can run. Use when the user says "set up growthbook", "configure my api key", "growthbook isn't working", "where do I put my key", or when another skill emits an error pointing here ("GBAPIKEY is not set", "authentication failed"). Writes ~/.config/growthbook/.env with chmod 600 and validates against the live API. For listing flags or running experiments, the other skills handle that.

Search, list, and audit GrowthBook metrics and fact tables. Use when the user asks "what metrics do we have", "find our revenue metric", "what fact tables exist", "which metrics are official", "what can I chart", "show me metrics tagged growth", "what columns does the orders fact table have", or "audit our metrics". Read-only — for actually charting a metric, use analytics-explore. For designing an experiment around a metric, use experiment-design.