ClariLayer — Claude Code MCP server: memory and setup
ClariLayer saves selected decisions and work preferences for AI to recall across projects and sessions. Connecting the server alone does not ensure the client will use that context.
claude mcp add --transport http clarilayer https://clarilayer.com/api/mcp/mcpWhat ClariLayer does
ClariLayer stores the facts, rules, decisions, and lessons you choose to carry forward. The AI can search an authorized space with recall_context, save confirmed information with remember, and open a full record with get_context_entry. To change or remove an entry, first retrieve it and use its ID.
Connect over HTTP, then finish authentication in your client. At the first relevant use in a new session, the installed setup can retrieve current workflow guidance. MCP connection alone does not make the AI consult saved context; the client has to call the server tool.
History import is a separate flow: review a preview before accepting it, and grant access separately for ongoing capture. The chosen extraction provider processes imported material; ClariLayer keeps extracted context rather than raw transcript archives. Semantic search needs separate organization consent, while lexical recall remains available without it.
What it helps with
- carry a confirmed decision into a later session
- retrieve project context before drafting a page
- save a durable work preference or rule
- check the source and scope of a saved entry
- correct or remove a specific memory
Example prompts for the agent
- Recall the Launch project's decisions before drafting the page, and show each source.
- Save this confirmed project rule and identify where it applies.
- Find the Launch decision entry and remove it using its ID.
How to install ClariLayer
Platform CLI: remote server
claude mcp add --transport http clarilayer https://clarilayer.com/api/mcp/mcpPlatform CLI: remote server
codex mcp add clarilayer --url https://clarilayer.com/api/mcp/mcpConfig file
{
"mcpServers": {
"clarilayer": {
"url": "https://clarilayer.com/api/mcp/mcp"
}
}
}Config file
{
"servers": {
"clarilayer": {
"type": "http",
"url": "https://clarilayer.com/api/mcp/mcp"
}
}
}Platform CLI: remote server
gemini mcp add -t http clarilayer https://clarilayer.com/api/mcp/mcpStep by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about ClariLayer
How do I install ClariLayer in Claude Code?
Run claude mcp add --transport http clarilayer https://clarilayer.com/api/mcp/mcp, then complete authentication in the client.
What does ClariLayer do?
It stores selected work context and lets an AI retrieve it from an authorized space. Saved entries retain their source and applicability.
Does connecting ClariLayer import conversation history?
No. History import is a separate process with source and provider selection, a preview, and approval. Connecting MCP does not read history.
