Lunge MCP server
io.github.thuupx/lunge·v1.3.1Agent-native API client: execute/test REST/GraphQL/WS/SSE with assertions, extraction, collections
0 stars24 downloads/wk
Reviews
Write oneNobody has reviewed Lunge yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Lunge tools (22, 3 write)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
collection_formatReturn the Lunge collection format reference: top-level fields, step schema,
graphql_introspectIntrospect GraphQL endpoint, return summarized schema with field signatures (args + return types) and input types.
graphql_requestGraphQL query/mutation over HTTP. Supports {{var}} extraction.
http_requestREST/HTTP request with auth, assertions, {{var}} extraction. Returns token-efficient summary.
import_curlParse curl into structured request. Optionally execute it.
import_harParse HAR file → Lunge collection. Splits query/headers/body. Optional 2xx filter.
import_openapiParse OpenAPI/Swagger (JSON/YAML) → Lunge collection. One step per operation with 2xx assertion.
inspect_responseDrill into a stored response by handle without re-running the request.
list_collectionsFind collection files (.yaml/.yml/.json) under a directory.
list_envsList env names and (masked) variables of the active or specified one.
list_historyList all requests recorded in this session with full specs (method, url, query, headers, body, auth, assert, extract)
run_collectionwrite actionRun declarative collection file. Threads extracted vars between steps. Optional JUnit XML export.
save_collectionSave multiple requests as a multi-step collection file. Params:
save_requestPersist last ad-hoc request (or explicit spec) as a new step in a collection file.
set_envwrite actionCreate/update named env. Values may be {value, secret:true}. Use via {{name}}.
set_policyConfigure host allow/deny lists + dry-run mode. Glob patterns: *, *.example.com.
sse_sessionSubscribe to SSE, collect events until stop condition. Returns summarized batch with assertions.
ws_closeClose a persistent WebSocket connection and release its handle.
ws_openOpen long-lived WS, return handle. Use ws_send/ws_recv/ws_close to interact.
ws_recvDrain buffered frames from persistent WS. Blocks up to maxDurationMs for first frame.
ws_sendwrite actionSend a single message on a persistent WebSocket connection opened with ws_open.
ws_sessionConnect WS, send messages, collect frames until stop condition. Returns summarized batch.
Public scan report
scanner v0.1.9 · 2026-09-25 · same rubric, same numbers if you re-run it
- Code scan41 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitylocal package, no credentials required12/15
- Maintenancelast push 65 days ago12/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
How Lunge compares
Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 88/100
- higher than 80% of 4,460 AI and agents servers
- Tools
- 22
- the typical AI and agents server has 6
- Tools that change things
- 14%
- typical: 0%. Fewer write tools means less an agent can break
- Downloads a week
- 24
- more than 10% of 1,538 AI and agents servers with a package
What its tools let an agent do: 13 read · 5 create · 3 change · 1 run or deploy
Install Lunge in Claude Code, Cursor or VS Code
Runs npx -y lunge-mcp on your machine. Read the scan report first; the gateway never runs local packages.
claude mcp add lunge -- npx -y lunge-mcp
What the publisher says
From the Lunge repository's README, as published. We do not edit it. Read it on GitHub
lunge
An agent-native advanced API client - execute and test REST, GraphQL, WebSocket, SSE, and (later) gRPC requests from any MCP-capable AI agent. No GUI, no manual clicking, no heavyweight desktop app. Think advanced curl, native for agents.
It is a lightweight, token-efficient alternative to Postman/Bruno for the era of AI coding agents. The agent calls tools directly; there are no humans in the loop and no bloated feature surface to pay for.
Why this exists
Design pillars
payloads are spilled to disk and referenced by handle so the context stays small.
- Agent-native - designed for LLM tool calls, not human UIs.
- Token-efficient - every response is filtered, truncated, and summarized. Full
a thin TypeScript MCP layer via napi-rs.
- Fast & portable - a Rust core (execution + protocols + summarization) exposed to
re-run declarative collections stored as files.
- Both ad-hoc and reusable - agents can fire one-off requests or create, save, and
Documents
Decisions locked in
- Stack: Rust core via napi-rs + TypeScript MCP layer (@modelcontextprotocol/sdk).
- Distribution: local dev tool over stdio transport (used by Cursor/Windsurf/Claude Desktop).
- Test definition: both ad-hoc agent calls and persisted declarative collections.
- MVP protocols: REST, GraphQL, WebSocket, SSE. gRPC is a later phase.
Nothing above is checked by us. What we check is on the safety report.
Lunge: common questions
- Is Lunge MCP server safe?
- Yes, by our scan: it is graded A (88/100). Read the Lunge safety report
- How do I install Lunge?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Lunge need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is Lunge maintained?
- The last commit was 104 days ago (2026-07-23). The latest release is v1.3.1.
- What can I use instead of Lunge?
- Servers from other publishers that do the same job: Mockserver MCP server, FireCMS Cloud MCP server and Agent-Native Slides MCP server.
- How does Lunge compare with other AI and agents MCP servers?
- safety score 88/100 (higher than 80% of 4,460 AI and agents servers); tools 22 (the typical AI and agents server has 6); tools that change things 14% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to Lunge
Same job from other publishers: the closest match first, then the best rated.
- MockserverMock, record/replay, verify and chaos-test any HTTP, REST, gRPC or LLM dependency over MCP.not reviewedEstablishedA
- FireCMS CloudManage FireCMS Cloud from an AI agent: Firestore data, collections, schemas and users.not reviewedEstablishedA
- Agent-Native SlidesAgent-Native Google Slides - generate and edit React presentationsnot reviewedEstablishedA
- Ts CoreAgent-native TypeScript framework for building MCP servers.not reviewedEstablishedA
- You.com Web Access & AIWeb search, AI agent, and content extraction via You.com APIsnot reviewedEstablishedA