An MCP server that indexes a codebase structurally and exposes tools for finding symbols, source, dependencies, dependents, change impact, and related project metadata. It supports many programming and config file types and is designed to return small, targeted context instead of whole files.
From the registry: Structural codebase MCP server — navigate and edit code by symbol name, not raw text.
$ python3 --versionPlease install the `token-savior` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Python** — Python 3.11+ Run: `python3 --version`
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["token-savior"]
- optional environment variables:
- `WORKSPACE_ROOTS`: Comma-separated list of project root paths to index (example: `/path/to/project1,/path/to/project2`)
- `TOKEN_SAVIOR_CLIENT`: Client name for live dashboard attribution (example: `claude-code`)
- `TOKEN_SAVIOR_MAX_FILES`: Max files per project (default 10000) (example: `10000`)
- `TOKEN_SAVIOR_MAX_FILE_SIZE_MB`: Max file size in MB (default 1) (example: `1`)
Note: 53 tools for structural codebase indexing. Supports Python, TS/JS, Go, Rust, C#, C/GLSL. 97% token reduction measured in real sessions.
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.Real-time Kubernetes network traffic visibility and API analysis for HTTP, gRPC, Redis, Kafka, DNS.