"Actual Budget" matching MCP servers:

Enables Claude Code to delegate bounded repository tasks to DeepSeek as a local sub-agent, handling exploration, routine changes, and test runs within a controlled workspace and budget.

Enables AI coding agents to set budgets per task, check costs before expensive operations, and halt when budget is exhausted, with support for calibration and cloud sync.

Enables Claude Code to research design references per page section across galleries like Awwwards, Dribbble and Mobbin, render the user's page in multiple style directions with live 3D components, and collect votes on a local board. It then produces a PDF report and iteratively builds and refines the actual page, dashboard prototype or mobile screens from that feedback.

Indexes GitHub repositories and provides living documentation context to coding agents, enabling them to ground work in actual repository conventions and propose doc updates.

Enables web AI agents to operate local projects through MCP, with real-time display of the actual model name and thinking effort.

Transforms single-attempt coding into a multi-attempt, test-validated refinement loop by running your actual test suite and feeding failures back to the LLM as structured directives.

Splits AI coding workflows into plan, implement, review, and evaluate steps, routing each to a different AI subscription to manage costs and prevent budget overuse.