9 tools from the TokenOracle MCP Server, categorised by risk level.
View the TokenOracle policy →budget_check Checks whether a planned LLM task fits within a monthly budget. Accepts prompt_text, task descriptions, or explicit token counts. Returns a go/no-g... compare_models Compares LLM pricing across all supported providers for a given task. Accepts prompt_text, task descriptions, or explicit token counts. Returns onl... estimate_cost Estimates the USD cost of an LLM API call BEFORE you make it. Does NOT make any LLM API call itself. Provide prompt_text, a task description, or ex... estimate_cost_batch Estimates costs for multiple LLM tasks in a single call. Each task can provide prompt_text, a task description, or explicit token counts. Returns p... find_cheapest_for_budget Inverse of budget_check. Given a budget cap and task, returns the best context-compatible model/quality combination you can afford. Accepts prompt_... get_budget_status Returns current-month tenant and actor spend totals from persisted request history. When the tenant has a configured monthly budget, also returns r... get_usage_leaderboard Returns a tenant-scoped leaderboard of the biggest spend drivers over a trailing window, ranked by actor by default or by tool when requested. get_usage_summary Returns aggregated request usage over a trailing time window for the authenticated tenant by default, or the caller actor when scope=actor. Include... list_request_activity Returns recent persisted request activity for the authenticated tenant by default, or for the caller actor when scope=actor. Includes metered cost ... The TokenOracle MCP server exposes 9 tools across 1 categories: Read.
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 TokenOracle server.
TokenOracle tools are categorised as Read (9). Each category has a recommended default policy.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept