Prism Coder runs on your own machine — the client starts it, so there is no endpoint to ping. 1 739 installs a week from npm. Last commit 15 Oct 2026.
Session memory for coding agents: local-first recall, drift detection, on-device inference.
We read the source, 19 h ago · rules ccca72095570
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
postinstall: node -e "import('./dist/postinstall.js').catch(()=>{})"
result = subprocess.run(
const result = spawnSync(
const vaultPath = path.join(mediaDir, filename);
const entries = await readdir(root, { withFileTypes: true }).catch(() => []);
const pngBase64 = "iVBORw0KGgoAAAANSUhEUgAAALQAAAC0CAYAAAA9zQYyAAAAAXNSR0IArs4c6QAAAERlWElmTU0AKgAAAAgAAYdpAAQAAAABAAAAGgAAAAAAA6ABAAMAAAABAAEAAKACAAQAAAABAAAAtKADAAQAAAABAAAAtAAAAABW1ZZ5AAAch0lEQVR4Ae2dXZsdx1HHZ3e1K620u/K7LVmO47eEx5YMGBuHGyCQ8BYuA4EAV5CvwZfgm3DPV+COO665g…
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add prism-coder -- npx -y prism-mcp-server
{
"mcpServers": {
"prism-coder": {
"args": [
"-y",
"prism-mcp-server"
],
"command": "npx"
}
}
}
[mcp_servers.prism-coder]
command = "npx"
args = ["-y", "prism-mcp-server"]
{
"mcpServers": {
"prism-coder": {
"args": [
"-y",
"prism-mcp-server"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"prism-coder": {
"args": [
"-y",
"prism-mcp-server"
],
"command": "npx"
}
}
}
This one needs environment variables set before it will start:
BRAVE_API_KEY (Brave Search API key for web search tools), GEMINI_API_KEY (Google Gemini API key for AI-powered summarization and embeddings), SUPABASE_URL (Supabase project URL for session memory and knowledge persistence), SUPABASE_KEY (Supabase service role key for database access), PRISM_USER_ID (User ID for multi-tenant row-level security isolation (defaults to 'default')), GCP_PROJECT_ID (Google Cloud project ID for Vertex AI Discovery Engine integration), VERTEX_LOCATION (Vertex AI location/region (defaults to 'global')), VERTEX_DATA_STORE_ID (Vertex AI Discovery Engine data store ID for enterprise search).
The author declared them in the registry entry; get the values from the project itself.
Local-first memory for AI coding agents. Memory + CodeGraph + Wiki in one SQLite file.
Deterministic context selection for AI coding agents. Local memory that compounds across sessions.
Local-first codebase intelligence MCP server for AI coding agents.
Markdown memory for coding agents that audits itself: drift reports and session digests.
Local code-memory graph for AI coding agents.
Local-first persistent session memory for Claude Code and other agents. SQLite + FTS5, no cloud.
Shared local memory for coding agents: SQLite, semantic recall, linked entities, code navigation.
Local-first memory for AI coding agents: shell history with exit codes, git diffs, docs, MCP.
Answers built from our own checks of this server.