Tokenadevsmcp-codebase-index

Best alternative to mcp-codebase-index

Tokenade is the best alternative to mcp-codebase-index — Universal token-optimization engine for AI coding agents — native hooks for 18 agents combine output filtering, semantic code search, skeleton compression, sandboxed execution, MCP proxying and a live savings dashboard in a single dependency-free binary.

Get Tokenade
Tokenademcp-codebase-index

Output Filtering

Format-aware compactors cover git, cargo, kubectl, terraform, docker and more — 60–99% reduction on the noisiest commands. Command rewriting further trims source-side before the shell even runs.

Output Filtering

Not available. mcp-codebase-index focuses on structural code queries; it does not filter CLI command outputs.

Semantic Code Search

Finds the most relevant files for a task and sends only those to the model, instead of the whole repo. Runs fully on-device with no external vector database and no model downloads — fast even on large codebases.

Structural Code Indexing

AST-based indexing for Python (stdlib ast) and regex-based for TypeScript/JS, Go, Rust, C# and Markdown. 18 query tools with sub-millisecond symbol lookup and transitive dependency analysis. Response size scales with the answer, not the codebase.

Skeleton Compression

Signatures-only view of source files, YAML, Markdown and Terraform — −64% on file reads while preserving every top-level declaration. Stacks on top of output filtering for maximum savings.

Persistent Cache

Persistent .codebase-index-cache.pkl with git-diff incremental updates avoids full rebuilds on small changes. Token budget controls (max_results, max_lines) keep response sizes bounded.

Third-Party MCP Optimization

tokenade mcp-proxy wraps any third-party MCP server's launch command in the agent's MCP config, so every tool result (verbose JSON, logs, console output) is folded on the way back — set once, not per call. Image results pass through untouched.

MCP Tool Set

18 MCP tools. No manifest lazy-loading; all tools exposed upfront. The usage_stats tool provides session efficiency metrics (tool calls vs chars returned vs total source).

Mechanism Breadth

The only tool combining output filtering + semantic search + skeleton compression + sandbox execution + MCP proxying + secret redaction + content-addressed cache in a single binary. On the open THOL benchmark (Claude Code 2.1.183 campaign) it is the only tool measured significantly cheaper than the control: cost ratio 0.84 [0.72, 0.95], 120/120 successful runs.

Mechanism Breadth

Structural indexing + dependency graph + transitive impact analysis + persistent cache. Uniquely deep on change-impact analysis; no output filtering, semantic vector search or skeleton compression.

Savings Dashboard

tokenade dashboard shows measured savings, per-command and per-project breakdown, and framework-detection status. Local logs rotate automatically with built-in secret redaction.

Session Efficiency Stats

The usage_stats MCP tool reports session efficiency: total tool calls, total chars returned, and total source size — a lightweight indicator of how much context the index saved.

Pricing

Freemium: works without an account (10M tokens offered per machine to try); a free account raises that to 10M tokens saved per month on unlimited machines. Pro at $19.90/mo (€19.90 incl. tax) covers 100M/month, then optional pay-as-you-save at $0.20 excl. tax (€0.20 incl. tax) per extra million saved. Enterprise on quote ([email protected]).

Pricing

Free under AGPL-3.0 for open-source use. Commercial use requires a separate license agreement.

mcp-codebase-index at a glance

mcp-codebase-index starts at Free (AGPL-3.0). Zero-dependency Python MCP server using stdlib AST for structural indexing (Python) and regex (TS/JS, Go, Rust, C#, Markdown), with 18 query tools, persistent disk cache, sub-millisecond symbol lookups and transitive dependency analysis.

Pros

  • Sub-millisecond symbol lookup (0.01–0.08 ms on CPython's 1.1M lines)
  • Persistent cache with git-diff incremental updates — only re-parses changed files
  • Change impact analysis with transitive dependents — unique to this tool
  • Zero dependencies: pure Python stdlib, no pip install for the core
  • Response size scales with the answer, not the codebase size

Cons

  • Limited language support: no PHP, Ruby, Java, C++ — only Python (AST) + 4 others (regex)
  • AGPL-3.0 license requires a commercial license for proprietary use
  • No routes/schema/framework detection (unlike codesight)
  • No output filtering, no semantic vector search, no skeleton compression

Ready to cut costs with Tokenade?

Join the teams that already chose Tokenade over mcp-codebase-index.

Get Tokenade

Other comparisons