echocache MCP
An MCP server that caches expensive LLM results using HTTP-style freshness and semantic recall. It supports exact-match caching based on model and prompt parameters, as well as knowledge-graph recall to find related entries by meaning. This allows agents to reuse previous conclusions, significantly reducing token usage for repeated research or judgment tasks.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Caching expensive LLM research chains to avoid re-deriving conclusions
- Semantic recall of related answers using a similarity graph
- Reducing token costs for re-orientation across sessions
Source and maintenance
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
echocache 0.1.3
{
"version": "0.1.3",
"transport": "stdio",
"identifier": "echocache",
"registryType": "npm"
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"echocache-mcp": {
"command": "npx",
"args": [
"-y",
"echocache@0.1.3"
]
}
}
}Cline
{
"mcpServers": {
"echocache-mcp": {
"command": "npx",
"args": [
"-y",
"echocache@0.1.3"
]
}
}
}VS Code
{
"servers": {
"echocache-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"echocache@0.1.3"
]
}
}
}Claude Desktop
{
"mcpServers": {
"echocache-mcp": {
"command": "npx",
"args": [
"-y",
"echocache@0.1.3"
]
}
}
}Let your agent find its tools
Find servers like echocache MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.