Updates
2026-04-15: added git hook auto-maintenance with--quickhealth check (260x speedup), stale index cleanup CLI, and three-layer hook protection (debounce + quick mode + daily throttle).
Potencia tu agente
Instalación del publicador. No es un endpoint de PlugMCP.
Host a copy with our keys and logs.
Aún no hay una versión empaquetada en el registro. Sigue el README del repositorio.
¿URL remota o un comando en tu máquina? En qué se diferencian las dos instalaciones
Ejecutar en PlugMCP
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
AbrirDel publicador
2026-04-15: added git hook auto-maintenance with --quick health check (260x speedup), stale index cleanup CLI, and three-layer hook protection (debounce + quick mode + daily throttle).ContextAtlas — context infrastructure for AI coding agents: hybrid retrieval, project memory and retrieval observability via CLI, MCP server or embeddable library. Tree-sitter indexing, LanceDB vector search, FTS5 and token-aware context packing.
2026-04-14: extracted 11 Memory MCP tools' business logic to application layer, completing three-layer architecture separation (CLI / MCP adapter → application → domain), and fixed all 31 TypeScript compilation errors.2026-04-10: codebase-retrieval now includes the lightweight direct graph summary by default, with MCP metadata, tests, and changelog docs updated in sync.2026-04-09: added churn / cost-aware index planning, moved long-term memory into dedicated tables + FTS5, and finished default-path hardening, threshold configuration, ops alert threshold alignment, and doc sync.2026-04-08: added the embedding gateway, local caching and multi-upstream routing, plus Hugging Face integration and MCP context lifecycle tools.2026-04-07: improved the indexing pipeline with lighter planning, snapshot copy reduction, queue observability, fallback hardening, and repeatable benchmarks.2026-04-06: tightened the default user path, memory governance, and operational visibility to make first use, feedback loops, and health checks clearer.ContextAtlas is not just a code search tool. It addresses a more practical engineering problem:
If you are building Claude Code workflows, MCP clients, or custom agent systems, ContextAtlas provides a context infrastructure layer: retrieval, memory, context packing, and observability.
In real projects, agent failures are often not caused by a weak model. They come from weak context systems:
ContextAtlas turns this into a composable set of capabilities:
ContextAtlas decides what context to provide, not how the task should be executed. It does not handle agent reasoning, workflow orchestration, or business API actions.
npm install -g @codefromkarl/context-atlas
Product identity mapping:
ContextAtlas@codefromkarl/context-atlascontextatlasAvailable commands:
contextatlascw (short alias)The docs use contextatlas as the primary command name. cw remains as a compatibility alias.
Initialize the config directory and example environment file first:
contextatlas init
# Choose your integration mode:
contextatlas setup:local --mode cli-skill # Terminal + skill integration
# OR
contextatlas setup:local --mode mcp # MCP client integration
Default config file location:
~/.contextatlas/.env
Minimum required configuration:
EMBEDDINGS_API_KEY=
EMBEDDINGS_BASE_URL=
EMBEDDINGS_MODEL=
RERANK_API_KEY=
RERANK_BASE_URL=
RERANK_MODEL=
Index update planning also supports these optional knobs:
INDEX_UPDATE_CHURN_THRESHOLD=0.35
INDEX_UPDATE_COST_RATIO_THRESHOLD=0.65
INDEX_UPDATE_MIN_FILES=8
INDEX_UPDATE_MIN_CHANGED_FILES=5
INDEX_UPDATE_CHURN_THRESHOLD: when the changed-file ratio crosses this value, index:plan / index:update will favor fullINDEX_UPDATE_COST_RATIO_THRESHOLD: triggers full when the estimated incremental cost is close to a full rebuildINDEX_UPDATE_MIN_FILES / INDEX_UPDATE_MIN_CHANGED_FILES: require both repo size and change size to clear a minimum bar before escalation is allowed
initwrites an editable example.env, including default SiliconFlow endpoints and recommended model settings.setup:local --mode <mode>writes only the configuration files for the selected mode. See First use guide for mode selection guidance. After setup, runcontextatlas health:fullto verify index, memory, graph, contract, and MCP process health.
If you are onboarding for the first time, start with the First use guide.
contextatlas start /path/to/repo
contextatlas init
# edit ~/.contextatlas/.env
contextatlas index /path/to/repo
contextatlas search \
--repo-path /path/to/repo \
--information-request "How is the authentication flow implemented?"
contextatlas daemon start
contextatlas mcp
If you want MCP client integration, run
contextatlas setup:local --mode mcpfirst.
Set up with contextatlas setup:local --mode cli-skill.
Useful for:
Example:
# retrieval
contextatlas search --repo-path /path/to/repo --information-request "Where is the payment retry policy implemented?"
# project memory
contextatlas memory:find "search"
contextatlas decision:list
# health
contextatlas health:full
Set up with contextatlas setup:local --mode mcp.
Use contextatlas setup:local --mode mcp --toolset retrieval-only when the client should only see read-only retrieval, graph, contract, and memory-reader tools.
Useful for:
Claude Desktop configuration example:
{
"mcpServers": {
"contextatlas": {
"command": "contextatlas",
"args": ["mcp"]
}
}
}
ContextAtlas MCP tools cover:
Indexing: Crawler / Scanner → Chunking → Indexing → Vector / SQLite Storage
Retrieval: Vector + FTS Recall → RRF → Rerank → Graph Expansion → Context Packing
Memory: Project Memory / Long-term Memory / Hub → CLI / MCP Tools
ContextAtlas focuses on what context to provide, not how the task should be executed. For a fuller architecture explanation, see repository positioning and engineering positioning.
Ways to improve ContextAtlas:
Before submitting code, it helps to:
pnpm build and make sure the repo still buildspnpm build
pnpm build:release
pnpm dev
node dist/index.js
MIT
ContextAtlas — context infrastructure for AI coding agents: hybrid retrieval, project memory and retrieval observability via CLI, MCP server or embeddable library. Tree-sitter indexing, LanceDB vector
Resumen general
ContextAtlas — context infrastructure for AI coding agents: hybrid retrieval, project memory and retrieval observability via CLI, MCP server or embeddable library. Tree-sitter indexing, LanceDB vector search, FTS5 and token-aware context packing.
Herramientas
Aún no hay lista de herramientas. Los servidores remotos la reciben tras la primera comprobación de estado.
Preguntas frecuentes
Copia en esta página la configuración stdio para Cursor, Claude Desktop, Claude Code o VS Code. Usa la instalación del publicador, no un proxy de PlugMCP.
Este registro no indica ninguna variable de entorno obligatoria.
ContextAtlas se ejecuta localmente por stdio.
Última comprobación: 2026-10-27. ContextAtlas — context infrastructure for AI coding agents: hybrid retrieval, project memory and retrieval observability via CLI, MCP server or embeddable library. Tree-sitter indexing, LanceDB vector search, FTS5 and token-aware context packing.