Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@suplagentics-mcp-server Scan for unimported opencode sessions."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A standalone Model Context Protocol server exposing SuplAgentics's real capabilities — Import History today, more over time — to any MCP-capable client (opencode, Claude Code, Claude Desktop), not just one dashboard.
SuplAgentics is a set of tools built on top of opencode and opencode-mem for importing existing Claude Code / opencode session history into a searchable memory store, and surfacing improvement suggestions mined from that history. This package is the portable capability layer — the same logic used by the SuplAgentics dashboard (an OpenChamber-based web UI), but callable directly by any MCP client.
Related MCP server: Central Brain
npm install -g suplagentics-mcp-serverAdd to ~/.config/opencode/opencode.json (or your project's own config for project-only scope):
{
"mcp": {
"suplagentics": {
"type": "local",
"command": ["node", "/path/to/suplagentics-mcp-server/src/index.js"],
"enabled": true
}
}
}A user-scope entry (the example above) applies to every project automatically — no per-project
setup needed.
suplagentics_import_scan — scans this machine for opencode session history (opencode.db)
and Claude Code session history (~/.claude/projects/) not yet imported into opencode-mem.
Returns real per-project session/memory counts and import status.
More tools are on the way as more of the underlying capability gets exposed directly.
MIT
Scans this machine for opencode session history (from opencode.db) and Claude Code session history (from ~/.claude/projects/) that has not yet been imported into opencode-mem. Returns real per-project session/memory counts and import status — the same data the SuplAgentics dashboard's own Import History page shows.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden: it discloses the two data sources it reads, that it reports per-project session/memory counts and import status, and frames the output against the dashboard's Import History page. It still omits cost/latency, whether the scan is purely read-only, and any filesystem permission requirements.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, front-loaded with the action and sources, then the return value. Dense and mostly waste-free, though the closing dashboard-comparison clause is more reassuring than strictly load-bearing.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description must describe returns, and it does (per-project session/memory counts plus import status). With zero parameters and no annotations, the main remaining gap is operational context such as read-only safety and scan cost, which keeps it short of complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool takes zero parameters, so the baseline is 4. The description correctly implies there is nothing to configure, though it does not explicitly say the scan is unconditional (whole machine).
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (scans) and two concrete resources (opencode.db session history and ~/.claude/projects/ Claude Code history), and narrows scope to history not yet imported into opencode-mem. An agent can tell exactly what this tool produces without opening anything else.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The 'has not yet been imported' qualifier implies the use case (checking what is left to import before running an import), but there is no explicit when-to-use/when-not statement or alternative named. With no sibling tools, the routing burden is low, so this is adequate rather than strong.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0suplagentics_import_scanScored across 1 tool
With only one tool, there is no possibility of overlap or confusion; its purpose is singular and clearly stated.
The single tool follows a clear verb_noun pattern (import_scan) with a consistent namespace prefix, and there are no other tools to create inconsistency.
A single tool is too thin for a server that appears to manage session import and memory operations; it likely needs companion tools for querying, managing, or configuring imports.
The surface only covers scanning and import status; obvious gaps include actually importing sessions, listing projects, or managing memory, leaving agents with no follow-up actions after a scan.
Persistent AI memory shared across Claude, ChatGPT, coding agents, and compatible MCP clients.
Persistent memory and cross-session learning for AI coding assistants (hosted remote MCP).
Private persistent memory for Claude, ChatGPT & Gemini via MCP - semantic search, zero-code setup.
Persistent memory for AI assistants. Save once; recall from Claude, ChatGPT, or any MCP client.
