Lambe
Let your coding agent slice and transform JSON, YAML, TOML, HCL, CSV, TSV, and Markdown in one pipeline instead of ad-hoc scripts.
Overview
Lambe is a MCP server for the Build phase that queries JSON, YAML, TOML, HCL, CSV, TSV, and Markdown through one composable pipeline syntax.
What is this MCP server?
- Single pipeline syntax across JSON, YAML, TOML, HCL, CSV, TSV, and Markdown
- Composable filters and transforms for agent tool calls over local or fetched documents
- stdio MCP transport with mcpb binaries for Linux and macOS (x64 and arm64)
- Server version 0.8.0 in the official MCP server schema
- Useful for inspecting lockfiles, env templates, and API payloads without a separate jq/yq step
- Supports 7 document types: JSON, YAML, TOML, HCL, CSV, TSV, Markdown
- 4 platform mcpb packages (Linux x64/arm64, macOS x64/arm64)
Community signal: 1 GitHub stars.
What problem does it solve?
Agents waste context and tokens juggling different CLI tools for every config and data format in a repo.
Who is it for?
Solo builders who want agents to inspect configs, exports, and API samples without custom parsing scripts.
Skip if: Teams that need a hosted SQL warehouse, real-time streaming analytics, or heavy binary document parsing.
What do I get? / Deliverables
After you register Lambe, your agent can run consistent pipelines across seven common formats from a single MCP tool set.
- Filtered or transformed excerpts from supported structured formats
- Repeatable pipeline results your agent can cite in implementation steps
- Faster config and export inspection without extra CLI dependencies
Recommended MCP Servers
Journey fit
Structured config and data files are edited and inspected most often while wiring backends, agents, and third-party configs during the build phase. Composable query pipelines fit agent-driven integration work where agents read mixed formats from APIs, IaC, and exports without leaving the MCP tool surface.
How it compares
MCP data-query integration, not a full analytics warehouse or agent skill playbook.
Common Questions / FAQ
Who is Lambe for?
Lambe is for indie developers and agent users who regularly need to query JSON, YAML, TOML, HCL, CSV, TSV, or Markdown during coding and integration work.
When should I use Lambe?
Use Lambe when an MCP agent must filter, reshape, or summarize structured files or responses and you want one pipeline syntax instead of format-specific shell tools.
How do I add Lambe to my agent?
Add the Lambe MCP server via stdio using the mcpb release for your OS and architecture from the hakimjonas/lambe GitHub releases, then point Claude Code or Cursor at that server entry.