{
"$schema": "https://leanctx.dev/schema/addon-registry-v1.json",
"registry_version": 1,
"addons": [
{
"addon": {
"name": "headroom",
"display_name": "Headroom",
"description": "Reversible context compression for tool outputs, logs, files and RAG chunks — 60–95% fewer tokens, originals retrievable on demand. Installs on add (uv tool install) and wires its MCP server (headroom mcp serve) into the lean-ctx gateway.",
"author": "chopratejas",
"homepage": "https://github.com/chopratejas/headroom",
"license": "Apache-2.0",
"categories": ["compression"],
"integration": "compression",
"keywords": ["compression", "tokens", "rag", "reversible", "ccr", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "headroom",
"args": ["mcp", "serve"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] },
"install": {
"manager": "uv",
"package": "headroom-ai[mcp]",
"version": "0.28.0",
"bin": "headroom"
}
},
{
"addon": {
"name": "sophon",
"display_name": "Sophon",
"description": "Deterministic context compressor for MCP agents — shrinks prompts, conversation history, code digests, file deltas and shell output with zero ML at query time. Single Rust binary; installs on add (npm) and wires its MCP server (sophon serve) into the lean-ctx gateway.",
"author": "lacausecrypto",
"homepage": "https://github.com/lacausecrypto/mcp-sophon",
"license": "MIT",
"categories": ["compression"],
"integration": "compression",
"keywords": ["compression", "deterministic", "rust", "tokens", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "sophon",
"args": ["serve"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] },
"install": {
"manager": "npm",
"package": "mcp-sophon",
"version": "0.5.4",
"bin": "sophon"
}
},
{
"addon": {
"name": "rtk",
"display_name": "RTK (Rust Token Killer)",
"description": "Single-binary CLI proxy that filters and compresses shell command output by 60–90%. Reaches MCP clients through the community rtk-mcp bridge (built from source), so the lean-ctx gateway can run commands through it.",
"author": "rtk-ai",
"homepage": "https://github.com/rtk-ai/rtk",
"categories": ["compression"],
"integration": "compression",
"keywords": ["compression", "shell", "cli", "tokens", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "repomix",
"display_name": "Repomix",
"description": "Packs a whole repository into a single AI-ready file. Its MCP server (npx repomix --mcp) lets agents pack local or remote repos and grep the output on demand.",
"author": "yamadashy",
"homepage": "https://repomix.com",
"license": "MIT",
"categories": ["code-intelligence"],
"integration": "codebase-pack",
"keywords": ["codebase", "packing", "context", "repo", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "repomix@1.16.0", "--mcp"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] }
},
{
"addon": {
"name": "serena",
"display_name": "Serena",
"description": "LSP-backed semantic code retrieval, editing and refactoring — the IDE for your coding agent. Connects over MCP via stdio or Streamable HTTP.",
"author": "oraios",
"homepage": "https://github.com/oraios/serena",
"license": "MIT",
"categories": ["code-intelligence"],
"integration": "code-symbols",
"keywords": ["lsp", "symbols", "refactor", "navigation", "semantic", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "uvx",
"args": ["--from", "serena-agent==1.5.3", "serena", "start-mcp-server", "--context", "ide"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] }
},
{
"addon": {
"name": "mcp-code-context",
"display_name": "MCP Code Context",
"description": "Compresses any codebase into LLM-ready semantic context with Tree-sitter WASM (TypeScript, JavaScript, Python, PHP, Dart) — symbol-level reads and AST-aware edits. Runs on add via npx; no install or native build needed.",
"author": "achatainga",
"homepage": "https://github.com/achatainga/mcp-code-context",
"license": "MIT",
"categories": ["code-intelligence"],
"integration": "code-symbols",
"keywords": ["ast", "tree-sitter", "symbols", "codebase", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "mcp-code-context@3.7.1"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] }
},
{
"addon": {
"name": "codebase-intelligence",
"display_name": "Codebase Intelligence",
"description": "Parses your whole repo with Tree-sitter into a semantic graph of symbols, references and dependencies — symbol search, call graphs and impact analysis. 18 tools, 8 languages, zero infrastructure or keys. Runs on add via npx.",
"author": "g-tiwari",
"homepage": "https://github.com/g-tiwari/mcp-codebase-intelligence",
"license": "MIT",
"categories": ["code-intelligence"],
"integration": "code-graph",
"keywords": ["call-graph", "symbols", "references", "tree-sitter", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "mcp-codebase-intelligence@0.3.0"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] }
},
{
"addon": {
"name": "codecompress",
"display_name": "CodeCompress",
"description": "Local code-intelligence index that gives agents surgical access to symbols — project outline, get/expand symbol, hot-path and references — backed by a SQLite index. Installs on add (.NET global tool) and wires its MCP server.",
"author": "MCrank",
"homepage": "https://github.com/MCrank/code-compress",
"license": "MIT",
"categories": ["code-intelligence"],
"integration": "code-symbols",
"keywords": ["symbols", "index", "outline", "references", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "codecompress-server"
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] },
"install": {
"manager": "dotnet",
"package": "CodeCompress.Server",
"version": "0.15.0",
"bin": "codecompress-server"
}
},
{
"addon": {
"name": "graphify",
"display_name": "Graphify",
"description": "Turns code, docs, schemas and more into one queryable knowledge graph for agents. Ships an MCP server (stdio + HTTP), but serving needs a graph built by `graphify extract` (an LLM-backed step), so it is listed rather than auto-installed.",
"author": "safishamsi",
"homepage": "https://github.com/safishamsi/graphify",
"license": "MIT",
"categories": ["code-intelligence", "memory"],
"integration": "code-graph",
"keywords": ["knowledge-graph", "code", "graphrag", "ast", "memory", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "memgraph-ingester",
"display_name": "Memgraph Ingester",
"description": "Structure-aware RAG for 150+ languages: the memgraph-ingester CLI parses a codebase into a Memgraph code graph with project memory; this addon wires its MCP server (uvx memgraph-ingester-mcp) into the lean-ctx gateway for high-level graph queries. Needs a running Memgraph instance (e.g. Docker) plus a graph built with the ingester CLI — set MEMGRAPH_INGESTER_MCP_BOLT_URI to your Bolt endpoint.",
"author": "ousatov-ua",
"homepage": "https://github.com/ousatov-ua/memgraph-ingester",
"license": "MIT",
"categories": ["code-intelligence", "memory"],
"integration": "code-graph",
"keywords": ["knowledge-graph", "rag", "memgraph", "ast", "memory", "community", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "uvx",
"args": ["--from", "memgraph-ingester-mcp==0.6.6", "memgraph-ingester-mcp"],
"env": { "MEMGRAPH_INGESTER_MCP_BOLT_URI": "bolt://localhost:7687" }
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": ["MEMGRAPH_INGESTER_MCP_BOLT_URI", "MEMGRAPH_INGESTER_MCP_READ_ONLY"] }
},
{
"addon": {
"name": "qmd",
"display_name": "QMD (Query Markup Documents)",
"description": "On-device search engine for markdown notes, meeting transcripts, docs and knowledge bases — BM25 + vector search + LLM re-ranking, fully local via node-llama-cpp GGUF models. Register folders as collections (qmd collection add), then agents search them through its MCP server (qmd mcp) wired into the lean-ctx gateway.",
"author": "tobi",
"homepage": "https://github.com/tobi/qmd",
"license": "MIT",
"categories": ["search", "memory"],
"integration": "none",
"keywords": ["markdown", "notes", "semantic-search", "bm25", "local", "community", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "@tobilu/qmd@2.5.3", "mcp"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] }
},
{
"addon": {
"name": "claude-context",
"display_name": "Claude Context",
"description": "Semantic code search over an entire codebase (hybrid BM25 + dense vectors) via @zilliz/claude-context-mcp. Needs a Milvus/Zilliz vector database and an embedding API key to run, so it is listed rather than auto-installed.",
"author": "zilliztech",
"homepage": "https://github.com/zilliztech/claude-context",
"license": "MIT",
"categories": ["code-intelligence"],
"integration": "none",
"keywords": ["semantic-search", "codebase", "embeddings", "retrieval", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "context-mem",
"display_name": "Context-Mem",
"description": "Local long-term memory and context compression for coding agents — a markdown vault with hybrid BM25 + vector search and progressive disclosure, no API key required. Installs on add (npm) and wires its MCP server (context-mem serve).",
"author": "JubaKitiashvili",
"homepage": "https://github.com/jubakitiashvili/context-mem",
"license": "MIT",
"categories": ["memory"],
"integration": "memory",
"keywords": ["memory", "compression", "local", "recall", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "context-mem",
"args": ["serve"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] },
"install": {
"manager": "npm",
"package": "context-mem",
"version": "4.0.0",
"bin": "context-mem"
}
},
{
"addon": {
"name": "mem0",
"display_name": "Mem0",
"description": "Drop-in long-term memory for agents (vector + graph + key-value). Provides an official MCP server, but needs an API key/backend to run, so it is listed rather than auto-installed.",
"author": "mem0ai",
"homepage": "https://github.com/mem0ai/mem0",
"license": "Apache-2.0",
"categories": ["memory"],
"integration": "memory",
"keywords": ["memory", "personalization", "recall", "vector", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "cognee",
"display_name": "Cognee",
"description": "Builds a queryable knowledge graph plus vector store from your data (ECL / GraphRAG pipeline). Installs on add (uv tool install) and runs fully local by default (SQLite + LanceDB + Kuzu); set LLM_API_KEY so it can process. Wires its MCP server (cognee-mcp) into the lean-ctx gateway.",
"author": "topoteretes",
"homepage": "https://github.com/topoteretes/cognee",
"license": "Apache-2.0",
"categories": ["memory"],
"integration": "memory",
"keywords": ["memory", "knowledge-graph", "graphrag", "recall", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "cognee-mcp"
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": ["LLM_API_KEY"] },
"install": {
"manager": "uv",
"package": "cognee-mcp",
"version": "0.5.4",
"bin": "cognee-mcp"
}
},
{
"addon": {
"name": "letta",
"display_name": "Letta (MemGPT)",
"description": "Stateful agent runtime that manages its own memory like an operating system (core + archival tiers). Exposes MCP-compatible tools, but needs a running server + database, so it is listed rather than auto-installed.",
"author": "letta-ai",
"homepage": "https://github.com/letta-ai/letta",
"license": "Apache-2.0",
"categories": ["memory"],
"integration": "memory",
"keywords": ["memory", "agent-runtime", "stateful", "memgpt", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "graphiti",
"display_name": "Graphiti (Zep)",
"description": "Temporal knowledge-graph memory for agents — a bi-temporal entity/edge graph with hybrid semantic, keyword and graph search. Ships an MCP server, but needs a Neo4j/FalkorDB backend and an LLM API key, so it is listed rather than auto-installed.",
"author": "getzep",
"homepage": "https://github.com/getzep/graphiti",
"license": "Apache-2.0",
"categories": ["memory"],
"integration": "memory",
"keywords": ["memory", "knowledge-graph", "temporal", "graphrag", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "zep",
"display_name": "Zep",
"description": "Memory platform for AI agents — persistent threads, a knowledge graph and semantic recall via a managed cloud or self-hosted Community Edition. Reaches agents over MCP, but needs an account/API key, so it is listed rather than auto-installed.",
"author": "getzep",
"homepage": "https://github.com/getzep/zep",
"license": "Apache-2.0",
"categories": ["memory"],
"integration": "memory",
"keywords": ["memory", "recall", "knowledge-graph", "cloud", "competitor", "mcp"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "sequential-thinking",
"display_name": "Sequential Thinking",
"description": "Structured step-by-step reasoning tool — the official MCP reference server for breaking a hard problem into revisable thoughts.",
"author": "modelcontextprotocol",
"homepage": "https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking",
"license": "MIT",
"categories": ["reasoning", "workflow"],
"integration": "none",
"keywords": ["thinking", "reasoning", "planning", "chain-of-thought"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-sequential-thinking@2025.12.18"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] }
},
{
"addon": {
"name": "lmd",
"display_name": "Lean-MarkDown (LMD)",
"description": "Directive-driven Markdown for agent plans. Reusable macros, phase-isolation and single-sourcing keep plans token-lean and cache-safe — written hard against lean-ctx, inspired by MarkdownAI.",
"author": "dasTholo",
"homepage": "https://github.com/dasTholo/lean-ctx/tree/feat-lmd-v1",
"license": "Apache-2.0",
"categories": ["plans", "workflow"],
"integration": "none",
"keywords": ["markdown", "plans", "macros", "directives", "phase-isolation", "subagents"],
"min_lean_ctx": "3.8.0"
}
},
{
"addon": {
"name": "everything",
"display_name": "Everything (MCP smoke-test)",
"description": "Official MCP reference server that exercises every protocol feature — install it to verify your lean-ctx gateway can discover and call downstream tools.",
"author": "modelcontextprotocol",
"homepage": "https://github.com/modelcontextprotocol/servers/tree/main/src/everything",
"license": "MIT",
"categories": ["testing"],
"integration": "none",
"keywords": ["test", "reference", "diagnostics", "gateway"],
"min_lean_ctx": "3.8.0"
},
"mcp": {
"transport": "stdio",
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-everything@2026.1.26"]
},
"capabilities": { "network": "full", "filesystem": "read_write", "env": [] }
}
]
}