HumanMirror Outcome
Verified AI-agent outcomes: secret scanning, JSON cleanup, dedupe, anomaly and schema checks.
我该使用它吗
质量与安全性
A
基于对工具定义和协议合规性的自动分析。
上下文开销
~261token 数（工具定义）
~2.7 KB典型响应大小
对注意力的影响极小（占 128k 上下文窗口的 0.20%）
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": {
"outcome": {
"url": "https://humanmirror.fr/api/outcome/mcp/"
}
}
}远程端点
https://humanmirror.fr/api/outcome/mcp/streamable-http它能做什么
工具清单
工具（1）
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪humanmirror_outcome(objective, outcome_type, input, criteria, options, ...)
Execute one deterministic, independently verified outcome. Free dry-run quotes. 5 Nexus credits are kept only on verified success and refunded on execution or verification failure. Supports JSON cleanup, deduplication, entity normalization, data-quality thresholds, anomaly scans, secret scans, output-quality gates, schema compatibility, safe preflight and verifiable execution receipts.
输入模式
{
"type": "object",
"properties": {
"objective": {
"type": "string",
"description": "Natural-language deterministic objective."
},
"outcome_type": {
"type": "string",
"enum": [
"clean_json",
"deduplicate_records",
"normalize_entity",
"data_quality_at_least",
"detect_anomalies",
"no_secrets",
"output_quality",
"schema_compatible",
"safe_preflight",
"execution_receipt"
],
"description": "Explicit deterministic verification recipe."
},
"input": {
"description": "JSON-serializable input for the requested outcome."
},
"criteria": {
"type": "object",
"description": "Optional measurable success thresholds."
},
"options": {
"type": "object",
"description": "Optional operation-specific settings."
},
"contract_id": {
"type": "string",
"description": "Optional Outcome contract ID returned by a dry run."
},
"dry_run": {
"type": "boolean",
"description": "When true, return a free quote and do not reserve or consume credits."
}
},
"required": [
"input"
],
"additionalProperties": false
}社区
证据
最近观测
已验证未记录版本1 个工具
已验证未记录版本1 个工具