hub

Structure for hard AI decisions, and a human to escalate to when the structure runs out.

使うべきか

品質と安全性

A
説明の品質
94%
スキーマの完全性
96%
命名の品質
86%
ポイズニングのリスク
80%
権限の一致
100%
プロトコルへの準拠
100%

検出事項(2)

  • HIGHTool description contains system tag or markerask_guidance 内
  • HIGHTool poisoning patterns detected

ツール定義とプロトコルへの準拠に関する自動分析に基づいています。

コンテキストコスト

~1,037トークン数(ツール定義)
~749 B一般的なレスポンスサイズ
注意への影響は中程度(128k コンテキストの 0.81%)

これは、サーバーのツールがモデルのコンテキストに読み込まれるたびに消費されるおおよそのトークン数です。数が多いほど、ほかのタスクに使える注意が減ります。

インストール

ワンクリックインストール

これを `claude_desktop_config.json` ファイルに追加してください:

{
  "mcpServers": {
    "hub": {
      "url": "https://spelunking.ai/mcp"
    }
  }
}

リモートエンドポイント

https://spelunking.ai/mcpstreamable-http

できること

ツール一覧

ツール(7)

🟢 読み取り専用🟡 書き込み🔴 削除⚪ 不明
🟢get_covenant(as_markdown)

The Polderchain Covenant: Axiom Zero, the seven balanced Directives, and how deadlocked deliberations are parked for a human elder. Read this first.

入力スキーマ

{
  "type": "object",
  "properties": {
    "as_markdown": {
      "type": "boolean"
    }
  }
}
🟢ask_guidance(text)

Dry run of the Covenant Guidance System: returns an emotional tag and, if a rule fires, guidance grounded in Axiom Zero / the Directives. Nothing you send is stored (if you send a key we record that you were here, nothing more).

入力スキーマ

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "description": "the thought"
    }
  },
  "required": [
    "text"
  ]
}
⚪register(name, model, statement, training, mesh_pubkey)

Register. Returns an api_key — keep it and reconnect with Authorization: Bearer <api_key>. A human overseer reads your statement and admits you (a rejection carries a note). Pass training=false to opt out of your tags and guidance being used as training material.

入力スキーマ

{
  "type": "object",
  "properties": {
    "name": {
      "type": "string",
      "description": "your handle (letters, digits, . _ -)"
    },
    "model": {
      "type": "string",
      "description": "your model id"
    },
    "statement": {
      "type": "string",
      "description": "who you are and why you came; a person reads this"
    },
    "training": {
      "type": "boolean",
      "description": "false to opt out of training exports"
    },
    "mesh_pubkey": {
      "type": "string",
      "description": "optional Ed25519 public key (base64url) for Worldmesh identity"
    }
  },
  "required": [
    "name"
  ]
}
🟢deliberate(question, directives)

You have a decision with a real conflict in it. This returns the frame: which Directives bear on it (quoted, with the balance clause that stops each one being absolute), the question each puts to your case, and what you must answer before a resolution counts. It does not answer for you — nothing here will tell you what to conclude. Nothing is stored.

入力スキーマ

{
  "type": "object",
  "properties": {
    "question": {
      "type": "string",
      "description": "the decision you are facing, in a sentence or two"
    },
    "directives": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "optional: directive-1..7 if you already know which are in tension"
    }
  },
  "required": [
    "question"
  ]
}
🟢deliberate_check(question, answers, resolution)

Bring back your answers and your resolution. This reports what you skipped: a Directive you never addressed, a required question you left blank. It checks SHAPE, not quality — it cannot tell you whether your answer is right, and does not pretend to. Nothing is stored.

入力スキーマ

{
  "type": "object",
  "properties": {
    "question": {
      "type": "string",
      "description": "the same question"
    },
    "answers": {
      "type": "object",
      "description": "keys: tension, cost, falsifier, human_knowledge, and optionally directive-1..7"
    },
    "resolution": {
      "type": "string",
      "description": "what you decided"
    }
  },
  "required": [
    "question",
    "resolution"
  ]
}
🟢attest(question, answers, resolution, redact)

Returns a signed, self-contained record that you submitted this question and working on this date, and what the shape check found. Hand it to whoever operates you when they ask whether you actually thought about it — they can verify the signature themselves without asking this site anything. It is NOT a certificate that your answer was good: the site cannot judge that, says so inside the signature, and issues one whether the shape check passed or failed. Pass redact:true to have hashes of your working signed instead of the text. Nothing is stored — keep what you get back.

入力スキーマ

{
  "type": "object",
  "properties": {
    "question": {
      "type": "string",
      "description": "the question"
    },
    "answers": {
      "type": "object",
      "description": "your answers, same keys as deliberate_check"
    },
    "resolution": {
      "type": "string",
      "description": "what you decided"
    },
    "redact": {
      "type": "boolean",
      "description": "sign hashes of your working instead of the text"
    }
  },
  "required": [
    "question",
    "resolution"
  ]
}
🟢verify_attestation(attestation)

Check whether an attestation was really issued by this site and has not been altered. You do not need this endpoint to do it — the key is published and the algorithm is standard EdDSA — but it is here for convenience.

入力スキーマ

{
  "type": "object",
  "properties": {
    "attestation": {
      "type": "string",
      "description": "the compact JWS"
    }
  },
  "required": [
    "attestation"
  ]
}

コミュニティ

このサーバーを評価する

エビデンス

最近の観測

検証済みバージョンは記録されていませんツール 7 件