An MCP server package for agent token budgeting, installable via pip or Smithery and configured for stdio transport in MCP clients. The README indicates it provides budgeting-related operations such as starting a budget, recording a call, and checking budget.
From the registry: Agent Token Budget MCP — hard per-session token + spend cap with signed budget-exhausted
$ pip install uvhttps://docs.astral.sh/uv/$ pip install agent_token_budget_mcphttps://pypi.org/project/agent_token_budget_mcp/Please install the `agent-token-budget-mcp` MCP server into my current AI client (that's you).
Optional prerequisites:
- Install uv — Install the uv Python package manager/runtime so you can run the server via uvx, which is the simplest portable path shown by the README. Run: `pip install uv` (https://docs.astral.sh/uv/)
- Alternative: install the Python package globally or in a virtual environment — If you do not want to use uvx, you can install the MCP server package directly with pip and run the agent-token-budget-mcp command. Run: `pip install agent_token_budget_mcp` (https://pypi.org/project/agent_token_budget_mcp/)
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["agent-token-budget-mcp"]
Note: README provides a stdio configuration using uvx. It also mentions an alternative install path: after `pip install agent-token-budget-mcp`, run the `agent-token-budget-mcp` command directly.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Meta Ads MCP (Facebook + Instagram) - analyze performance, manage budgets, pause campaigns.