Vara MCP — Field Intelligence Scan Server
Governed external-environment scanning: signal harvest, Field Intel Reports, per-signal lineage.
我该使用它吗
质量与安全性
B
基于对工具定义和协议合规性的自动分析。
上下文开销
~1,017token 数（工具定义）
~1.6 KB典型响应大小
对注意力有中等影响（占 128k 上下文窗口的 0.79%）
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中：
{
"mcpServers": {
"vara-mcp-server": {
"url": "https://vara-mcp-server.vercel.app/api/mcp"
}
}
}远程端点
https://vara-mcp-server.vercel.app/api/mcpstreamable-http它能做什么
工具清单
工具（6）
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪vara_run_scan(keywords, active_planes, sweep_depth_hours, scan_label, enable_dual_track, ...)
Execute a full Real Vara scan using the operational dual-track, multi-timescale pipeline. This tool changes Vara runtime state.
输入模式
{
"type": "object",
"properties": {
"keywords": {
"anyOf": [
{
"items": {
"type": "string"
},
"type": "array"
},
{
"type": "null"
}
],
"default": null,
"title": "Keywords"
},
"active_planes": {
"anyOf": [
{
"items": {
"enum": [
"social",
"scientific",
"tech",
"adjacent_possible",
"economic",
"dark",
"geopolitical",
"persons",
"firecrawl"
],
"type": "string"
},
"type": "array"
},
{
"type": "null"
}
],
"default": null,
"title": "Active Planes"
},
"sweep_depth_hours": {
"default": 24,
"title": "Sweep Depth Hours",
"type": "integer"
},
"scan_label": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Scan Label"
},
"enable_dual_track": {
"default": true,
"title": "Enable Dual Track",
"type": "boolean"
},
"high_novelty_floor": {
"default": 0.15,
"title": "High Novelty Floor",
"type": "number"
},
"weak_novelty_floor": {
"default": 0.05,
"title": "Weak Novelty Floor",
"type": "number"
},
"enable_multi_timescale": {
"default": true,
"title": "Enable Multi Timescale",
"type": "boolean"
},
"use_firecrawl": {
"default": false,
"title": "Use Firecrawl",
"type": "boolean"
},
"generate_fir": {
"default": false,
"title": "Generate Fir",
"type": "boolean"
}
},
"title": "vara_run_scanArguments"
}🟢vara_list_scans(limit, since, label_contains)
List historical Vara scan metadata from the local archive.
输入模式
{
"type": "object",
"properties": {
"limit": {
"default": 20,
"title": "Limit",
"type": "integer"
},
"since": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Since"
},
"label_contains": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Label Contains"
}
},
"title": "vara_list_scansArguments"
}🟢vara_get_scan(scan_id, include_signals)
Retrieve a complete historical VaraScanReport by scan_id or short prefix.
输入模式
{
"type": "object",
"properties": {
"scan_id": {
"title": "Scan Id",
"type": "string"
},
"include_signals": {
"default": true,
"title": "Include Signals",
"type": "boolean"
}
},
"required": [
"scan_id"
],
"title": "vara_get_scanArguments"
}🟢vara_query_signals(plane, min_novelty, entity, since, until, ...)
Query the committed Vault corpus and optionally the current Veil hold.
输入模式
{
"type": "object",
"properties": {
"plane": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Plane"
},
"min_novelty": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"title": "Min Novelty"
},
"entity": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Entity"
},
"since": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Since"
},
"until": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Until"
},
"origin": {
"default": "any",
"enum": [
"passed",
"veil_promoted",
"any"
],
"title": "Origin",
"type": "string"
},
"canonized_only": {
"default": false,
"title": "Canonized Only",
"type": "boolean"
},
"include_veil_hold": {
"default": false,
"title": "Include Veil Hold",
"type": "boolean"
},
"limit": {
"default": 50,
"title": "Limit",
"type": "integer"
}
},
"title": "vara_query_signalsArguments"
}🟢vara_get_veil_state(plane, signal_id, include_trajectories)
Inspect current Veil hold entries and trajectories.
输入模式
{
"type": "object",
"properties": {
"plane": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Plane"
},
"signal_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Signal Id"
},
"include_trajectories": {
"default": true,
"title": "Include Trajectories",
"type": "boolean"
}
},
"title": "vara_get_veil_stateArguments"
}⚪vara_generate_fir(scan_id, series, baseline, format, next_cycle_hint)
Generate an Operator-Tier or lightweight Field Intel Report from a scan_id.
输入模式
{
"type": "object",
"properties": {
"scan_id": {
"title": "Scan Id",
"type": "string"
},
"series": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Series"
},
"baseline": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Baseline"
},
"format": {
"default": "operator_tier",
"enum": [
"operator_tier",
"lightweight"
],
"title": "Format",
"type": "string"
},
"next_cycle_hint": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Next Cycle Hint"
}
},
"required": [
"scan_id"
],
"title": "vara_generate_firArguments"
}社区
证据
最近观测
已验证未记录版本6 个工具