10 tools from the Gsquad MCP Server, categorised by risk level.
View the Gsquad policy →gsquad_ads_action_list List recent ads action runs (paginated). Optional filters: provider, status. gsquad_ads_action_status Get status of an ads action run by public_id (ULID). Requires ads:read or ads:write on the token. gsquad_google_ads_leads List ingested Google Ads leads (paginated). Filter by date range and campaign/ad group fields. gsquad_health Check GSquad API health endpoint. gsquad_knowledge_search Semantic search over ingested knowledge chunks (pgvector RAG). gsquad_lark_leads List Lark-ingested leads (paginated). Filter by record_date with from/to (YYYY-MM-DD). gsquad_meta_campaign_metrics List Meta Ads campaign metrics (raw rows). metrics_grain: all | daily | snapshot. Omit account_id/account_name if only one account exists. gsquad_meta_campaign_metrics_summary Aggregated Meta metrics for a date range. Default metrics_grain=daily (avoids mixing snapshots). group_by: none | day | campaign. gsquad_query Query GSquad Brain endpoint with text input (RAG-backed chunk retrieval). The Gsquad MCP server exposes 10 tools across 2 categories: Read, Write.
Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the Gsquad server.
Gsquad tools are categorised as Read (9), Write (1). Each category has a recommended default policy.
Deterministic policy on every MCP tool call. Per-identity grants. Full audit log.