
Lambë
Gives Claude a query language for inspecting and transforming structured data without guessing at the schema. You can ask it to extract fields from JSON, YAML, TOML, HCL, CSV, or Markdown, and it will tell you the shape at each pipe stage before executing. The explain operation shows you what formats the result can serialize to, and when a query produces something incompatible with your target format, it suggests bridging fragments. Useful when you're working with unfamiliar API responses, auditing configuration files, or need Claude to pull specific fields from data it hasn't seen before. The query language is intentionally bounded: no recursion or lambdas, which is the trade-off that makes static shape inference possible.
- GitHub stars
- 1
- tools
- 5
- transport
- STDIO
- last updated
- Sep 7, 2026
Tools
5 toolsPublic tool metadata for what this MCP can expose to an agent.
lambe_queryextract/filter/transform, with an optional `schema` parameter that validates data structurally before the query runs.
extract/filter/transform, with an optional `schema` parameter that validates data structurally before the query runs.
No parameter schema in public metadata yet.
lambe_print_shapeinspect unfamiliar data; returns a JSON Schema subset document.
inspect unfamiliar data; returns a JSON Schema subset document.
No parameter schema in public metadata yet.
lambe_checkvalidate data against a JSON Schema. Returns `{"ok": true}` or `{"ok": false, "error": "..."}` naming the disagreement path.
validate data against a JSON Schema. Returns `{"ok": true}` or `{"ok": false, "error": "..."}` naming the disagreement path.
No parameter schema in public metadata yet.
lambe_explaintrace a query statically (with or without data); returns a structured JSON report with shape-per-stage, warnings, and writability.
trace a query statically (with or without data); returns a structured JSON report with shape-per-stage, warnings, and writability.
No parameter schema in public metadata yet.
lambe_assertboolean assertion on a query result.
boolean assertion on a query result.
No parameter schema in public metadata yet.
Sponsors
Readme
Lambë
A query language for structured data that shows you what you're working with.
lam queries JSON, YAML, TOML, HCL, CSV, TSV, and Markdown. Unlike other query tools, it tells you what your query does before you run it — the shape at each pipe stage, which output formats can serialize the result, what would go wrong.
Use it when you don't already know the data: inspecting an unfamiliar API response, auditing a Helm chart, verifying a CI pipeline's assumptions, or asking an AI agent to extract something without guessing at the structure.
$ lam --to toml '.dependencies | keys' pubspec.yaml
Error: TOML output requires a map at the root, got list<string>.
Try appending one of:
| as(toml) # Wraps the list under a single-entry map (equivalent to `{items: .}`).
$ lam --to toml '.dependencies | keys | as(toml)' pubspec.yaml
items = ["rumil", "rumil_parsers", "rumil_expressions"]
Queries are bounded and always terminate. No recursion, no lambdas, no def. That's the tradeoff: Lambë doesn't try to be a programming language, so its shape inference, --explain, --schema, and error remediations all work.
Details
- Repository
- github.com/hakimjonas/lambe
- Categories
- Documents & Knowledge, Data & Analytics
- Package
https://github.com/hakimjonas/lambe/releases/download/v0.14.0/lam-mcp-linux-x64- Transport
- STDIO
- Registry
- MCP registry, active