GrowthBook tools in Cogny
Cogny's GrowthBook integration exposes 11 tools (11 read) to an AI agent over MCP. The same tools are available on every plan, from your own AI client or inside the Cogny app.
- Read (11): only reads data.
- Write (0): creates or changes something.
- Destructive (0): deletes, removes or replaces something.
Over MCP your own AI client asks before it runs a write. To take a tool away from your agent, block it in Settings → Integrations. Plans, approvals and limits.
All tools
| Tool | Kind | What it does |
|---|---|---|
list_experiments | Read | List experiments. |
get_experiment | Read | Get a single experiment by id. |
list_experiment_results | Read | Get the latest non-dimension results snapshot for every experiment matching the filters — scan a whole portfolio in one call (GrowthBook caps this at roughly the 1000 most recent experiments). |
get_experiment_results | Read | Get the latest computed results for one experiment. |
list_features | Read | List feature flags. |
get_feature | Read | Get a single feature flag by its key/id. |
list_projects | Read | List projects used to scope experiments, features and metrics. |
list_fact_metrics | Read | List fact-table-based metric definitions. |
list_metrics | Read | List legacy (non-fact-table) metric definitions. |
list_segments | Read | List saved audience segments used in experiment analysis. |
list_data_sources | Read | List connected data warehouses/sources. |
Exact input schemas come from tools/list once GrowthBook is connected; the free plan is enough to check. Missing something? Your agent can ask for it with request_feature.