All licenses are one-time purchases for commercial use. Personal and other non-commercial use is free: pip install jcodemunch-mcp. Using it at work takes one of the licenses below, paid once, no subscription. A jCodeMunch license includes every 1.x update, with no time limit. Choose a Munch Trio bundle (code + docs + data together)
or license any tool individually.
get_session_stats adds them up. Try it on your own repo during the 30 days and see for yourself. For teams, org-rollup totals the estimates across every seat. Free for 14 days, no key needed.
Commercial use for a single developer. The right license if you're an independent developer, freelancer, or sole contributor adding AI code assistance to your workflow. Choosing one code-retrieval tool for a team? Here's how to run your own with-and-without test.
The complete token-efficiency stack for one developer. jCodeMunch indexes your codebases and delivers symbol-accurate retrieval (functions, classes, methods) via MCP. jDocMunch indexes your documentation and delivers section-level search across Markdown, RST, Jupyter notebooks, HTML, OpenAPI specs, and more. jDataMunch indexes CSV, Excel, Parquet and JSONL data files, plus SQL query logs for usage, giving AI agents surgical access to your data.
Together they eliminate the file-read problem across all three layers: code, docs, and data. Index once per project, query in milliseconds for the rest of the project's lifetime.
jCodeMunch indexes your source code using tree-sitter AST parsing and exposes a set of MCP tools that let AI agents retrieve functions, classes, methods, constants, and more by symbol name, not by reading entire files.
jDocMunch indexes documentation files and exposes MCP tools for section-level search, so AI agents can find the exact paragraph or API spec entry they need without ingesting an entire knowledge base.
Pairs naturally with jCodeMunch: code questions go to the symbol index, documentation questions go to the section index. Together they cover the full context-window problem.
jDataMunch indexes CSV, Excel, Parquet and JSONL data files, plus SQL query logs for usage, and exposes 39 MCP tools by default, so AI agents can answer data questions precisely without loading entire files.
Pairs naturally with jCodeMunch and jDocMunch: code questions go to the symbol index, doc questions go to the section index, and data questions go to the data index. Together they cover the full context-window problem across all three layers.
All Builder licenses cover one named developer for commercial use. Need a team license? See Studio below.
Commercial use for up to 5 developers at the same organization. The right tier for an AI-enabled engineering team standardizing on efficient context retrieval. Buying for a team? Here’s how the 14-day evaluation works.
All three retrieval layers for your whole team. Every developer gets the same low-token-cost access to code symbols, documentation sections, and database schemas through Claude, Cursor, or any MCP-compatible client.
One bundle, all three tools, up to 5 seats. The cleanest way to standardize AI-assisted development across a small engineering team.
The same surgical symbol retrieval as the Builder license, extended to cover up to five developers at your organization. Ideal for teams where multiple engineers are working across shared codebases or using AI coding assistants daily.
Section-level documentation retrieval for up to five developers. Keep your team's AI agents answering doc questions precisely: whether your docs are Markdown wikis, RST reference manuals, Jupyter notebooks, or OpenAPI specs.
Schema-level data retrieval for up to five developers. Keep your team's AI agents answering data questions precisely across CSV, Excel, Parquet and JSONL data files.
Studio licenses cover up to 5 developers at the same organization. For larger teams, see Platform.
For organizations standardizing on structured AI retrieval across departments or larger engineering teams. Unlimited internal developers at a single organization.
The complete retrieval stack for an entire organization. Deploy jCodeMunch, jDocMunch, and jDataMunch across your engineering org (internal tooling, CI pipelines, shared MCP servers) under a single license.
One bundle covers all three tools, no per-seat counting, no recurring fees. The most cost-effective option for orgs standardizing on the full j*Munch suite.
Unlimited internal use of jCodeMunch across your organization. Deploy as a shared MCP server, integrate into CI/CD pipelines, or let every developer run their own instance: all under one license.
Unlimited internal use of jDocMunch across your organization. Index and serve documentation to every AI agent in your org (internal wikis, API specs, runbooks, knowledge bases) under a single license.
Unlimited internal use of jDataMunch across your organization. Index and serve data-file context to every AI agent in your org (CSV, Excel, Parquet and JSONL files, plus SQL query logs) under a single license.
Need enterprise terms, SLA, or a custom deployment arrangement? Contact us. Buying for your company? Here’s one page with the terms to forward to whoever approves it.
Every SKU with direct purchase links.
| Product | Tier | Seats | Price | |
|---|---|---|---|---|
| Munch Trio Bundle | Builder | 1 developer | $99 | Buy |
| Munch Trio Bundle | Studio | Up to 5 developers | $449 | Buy |
| Munch Trio Bundle | Platform | Org-wide | $2,499 | Buy |
| jCodeMunch only | Builder | 1 developer | $79 | Buy |
| jCodeMunch only | Studio | Up to 5 developers | $349 | Buy |
| jCodeMunch only | Platform | Org-wide | $1,999 | Buy |
| jDocMunch only | Builder | 1 developer | $29 | Buy |
| jDocMunch only | Studio | Up to 5 developers | $99 | Buy |
| jDocMunch only | Platform | Org-wide | $499 | Buy |
| jDataMunch only | Builder | 1 developer | $39 | Buy |
| jDataMunch only | Studio | Up to 5 developers | $149 | Buy |
| jDataMunch only | Platform | Org-wide | $499 | Buy |