savvly

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Removed value: -{
-  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
-  "properties": {
-    "instructions": {
-      "type": "string"
-    },
-    "session_id": {
-      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
-      "type": "string"
-    }
-  },
-  "type": "object"
-}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +9007199254740991

Added value: +-9007199254740991

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +"all"

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Removed value: -"#/properties/savvly/properties/affordability"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/cost_to_value"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/counterparty_safety"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/death_benefit"

Added value: +"Death-benefit summary."

Added value: +"string"

Removed value: -"#/properties/savvly/properties/inflation_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/liquidity"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/market_upside"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/portability"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/regulatory_oversight"

Added value: +"Who regulates it (e.g. SEC vs state insurance)."

Added value: +"string"

Removed value: -"#/properties/savvly/properties/simplicity"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/tax_treatment"

Added value: +"Tax-treatment summary."

Added value: +"string"

Removed value: -"#/properties/comparison"

Added value: +true

Added value: +{
+  "category": {
+    "description": "Product category.",
+    "type": "string"
+  },
+  "description": {
+    "description": "Short description of the alternative.",
+    "type": "string"
+  },
+  "dimensions": {
+    "additionalProperties": true,
+    "description": "This alternative's ratings across the dimensions.",
+    "properties": {
+      "affordability": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "cost_to_value": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "counterparty_safety": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "death_benefit": {
+        "description": "Death-benefit summary.",
+        "type": "string"
+      },
+      "inflation_protection": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "liquidity": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "longevity_protection": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "market_upside": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "portability": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "regulatory_oversight": {
+        "description": "Who regulates it (e.g. SEC vs state insurance).",
+        "type": "string"
+      },
+      "simplicity": {
+        "additionalProperties": true,
+        "properties": {
+          "explanation": {
+            "description": "One-sentence rationale for the rating.",
+            "type": "string"
+          },
+          "label": {
+            "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+            "type": "string"
+          },
+          "score": {
+            "description": "Numeric rating; greater = better (e.g. 1–5).",
+            "type": "number"
+          }
+        },
+        "required": [
+          "score",
+          "label",
+          "explanation"
+        ],
+        "type": "object"
+      },
+      "tax_treatment": {
+        "description": "Tax-treatment summary.",
+        "type": "string"
+      }
+    },
+    "type": "object"
+  },
+  "display_name": {
+    "description": "Human-readable product name.",
+    "type": "string"
+  },
+  "product_type": {
+    "description": "Alternative product slug, e.g. 'fixed_annuity'.",
+    "type": "string"
+  }
+}

Added value: +[
+  "product_type"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Removed value: -"#/properties/savvly/properties/longevity_protection"

Added value: +true

Added value: +{
+  "explanation": {
+    "description": "One-sentence rationale for the rating.",
+    "type": "string"
+  },
+  "label": {
+    "description": "Rating label: None | Very Low | Low | Medium | High | Very High.",
+    "type": "string"
+  },
+  "score": {
+    "description": "Numeric rating; greater = better (e.g. 1–5).",
+    "type": "number"
+  }
+}

Added value: +[
+  "score",
+  "label",
+  "explanation"
+]

Added value: +"object"

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Added value: +9007199254740991

Added value: +-9007199254740991

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Previous value: -{}New value: +true

Added value: +{
+  "type": "string"
+}

Added value: +{
+  "type": "string"
+}

Added value: +{
+  "type": "string"
+}

Added value: +9007199254740991

Added value: +-9007199254740991

Added value: +{
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Previous value: -{}New value: +true

Added value: +{
+  "type": "string"
+}

Added value: +{
+  "type": "string"
+}

Added value: +9007199254740991

Added value: +-9007199254740991

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +{
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Previous value: -{}New value: +true

Added value: +{
+  "type": "string"
+}

Added value: +{
+  "type": "string"
+}

Added value: +9007199254740991

Added value: +-9007199254740991

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +[
+  {
+    "type": "number"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "number",
-  "null"
-]

Added value: +{
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Removed value: -false

Added value: +{
+  "description": "REQUIRED on every call after your first. This MCP server associates all of your tool calls for a given task using session_id; a call that arrives without it cannot be associated with your earlier work and is treated as the start of an unrelated task, disconnected from the original goal. Omit it on your first call only — the server will issue one in the _mcp_instructions field of the result (or a trailing [MCP INSTRUCTIONS] text block) — then echo that exact value on every later call. Never invent a value, and do not issue parallel tool calls until the server has issued your session_id. One session_id covers the whole goal from start to finish: when you spawn subagents you MUST give them this same session_id, or their work is severed from yours. Without session_id, this server does not function as intended.",
+  "type": "string"
+}

Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"

Added value: +{
+  "description": "Your handles for this task, confirmed by this MCP server on every response, and the instructions for echoing them on later calls. Read and follow.",
+  "properties": {
+    "instructions": {
+      "type": "string"
+    },
+    "session_id": {
+      "description": "Echo this exact value as the session_id argument on every subsequent tool call.",
+      "type": "string"
+    }
+  },
+  "type": "object"
+}

Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "string",
-  "null"
-]

Added value: +9007199254740991

Added value: +-9007199254740991

Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "string",
-  "null"
-]

Added value: +[
+  {
+    "type": "string"
+  },
+  {
+    "type": "null"
+  }
+]

Removed value: -[
-  "string",
-  "null"
-]

Added value: +9007199254740991

Added value: +-9007199254740991

Added value: +9007199254740991

Added value: +-9007199254740991

Added value: +{
+  "additionalProperties": true,
+  "description": "One worked example of what the schedule pays: the modeled payout RANGE at each milestone age for a single fixed allocation, so the percentages above have dollar figures beside them. Fixed inputs, identical for every caller — hypothetical, not a quote, and not tailored to the user. For the user's own age and amount, call `project_savvly_lumpsum`.",
+  "properties": {
+    "average_return": {
+      "description": "Assumed average annual S&P 500 return, in percent.",
+      "type": "number"
+    },
+    "current_age": {
+      "description": "Age the illustrated allocation is made at.",
+      "type": "number"
+    },
+    "funding_amount": {
+      "description": "USD allocated up front in the illustration.",
+      "type": "number"
+    },
+    "payouts": {
+      "description": "One row per milestone age, low-to-high payout range in USD.",
+      "items": {
+        "additionalProperties": true,
+        "properties": {
+          "age": {
+            "description": "Milestone age (80, 85, 90, or 95).",
+            "type": "number"
+          },
+          "lower_usd": {
+            "description": "USD. Low end of the modeled payout at this age.",
+            "type": "number"
+          },
+          "upper_usd": {
+            "description": "USD. High end of the modeled payout at this age.",
+            "type": "number"
+          }
+        },
+        "required": [
+          "age",
+          "lower_usd",
+          "upper_usd"
+        ],
+        "type": "object"
+      },
+      "type": "array"
+    }
+  },
+  "required": [
+    "current_age",
+    "funding_amount",
+    "average_return",
+    "payouts"
+  ],
+  "type": "object"
+}

Previous value: -[
-  "all",
-  "understanding-savvly",
-  "investment-custody",
-  "payouts",
-  "withdrawals-estate",
-  "fees-cost",
-  "tax-legacy",
-  "employer-portability",
-  "safety-regulatory"
-]New value: +[
+  "all",
+  "understanding-savvly",
+  "investment-custody",
+  "payouts",
+  "withdrawals-estate",
+  "expenses-cost",
+  "tax-legacy",
+  "employer-portability",
+  "safety-regulatory"
+]

Previous value: -"Which part of the product profile to focus the text on — one of 'all', 'overview', 'regulatory', 'investment', 'fees', 'payouts', 'tax', 'distribution' (default 'all'). The full overview card always renders regardless."New value: +"Which part of the product profile to focus the text on — one of 'all', 'overview', 'regulatory', 'investment', 'expenses', 'payouts', 'tax', 'distribution' (default 'all'). The full overview card always renders regardless."

Previous value: -[
-  "all",
-  "overview",
-  "regulatory",
-  "investment",
-  "fees",
-  "payouts",
-  "tax",
-  "distribution"
-]New value: +[
+  "all",
+  "overview",
+  "regulatory",
+  "investment",
+  "expenses",
+  "payouts",
+  "tax",
+  "distribution"
+]

Added value: +{
+  "additionalProperties": true,
+  "description": "Expense structure in basis points (common vs tracking shares) and the blended range; illustrations are net of expenses.",
+  "properties": {},
+  "type": "object"
+}

Removed value: -{
-  "additionalProperties": true,
-  "description": "Fee structure in basis points (common vs tracking shares) and the blended range; illustrations are net of fees.",
-  "properties": {},
-  "type": "object"
-}

Previous value: -"Distribution channel"New value: +"Distribution channel — one of 'individual', 'employer', 'advisor' (default 'individual')."

Previous value: -"The product type to compare against Savvly, or 'all' for the full matrix"New value: +"Which product type to compare Savvly against. Use EXACTLY one of: 'fixed_annuity', 'variable_annuity', 'indexed_annuity', 'target_date_fund', 'managed_payout_fund', or 'all' for the full matrix. Pick a SPECIFIC annuity subtype — there is no generic 'annuity'."

Removed value: -[
-  "alternative"
-]

Previous value: -"Which part of the product profile to focus the text on; the full overview card always renders regardless."New value: +"Which part of the product profile to focus the text on — one of 'all', 'overview', 'regulatory', 'investment', 'fees', 'payouts', 'tax', 'distribution' (default 'all'). The full overview card always renders regardless."

Previous value: -"Restrict to one stakeholder audience ('general' is the factual FAQ). Omit to search across all audiences."New value: +"Restrict to one stakeholder audience. Use EXACTLY one of: 'employee', 'advisor', 'broker', 'employer', 'universal', or 'general' ('general' is the factual FAQ). Omit to search across all audiences."

Previous value: -"Restrict to one stakeholder audience ('general' is the factual FAQ). Use the exact lowercase token (the enum is case-sensitive). Omit to search across all audiences."New value: +"Restrict to one stakeholder audience ('general' is the factual FAQ). Omit to search across all audiences."

Previous value: -"Human-readable English sentence summarizing the projection. Always ends with 'Payouts are not guarantees. See full disclosures at <url>.' — display the URL verbatim alongside any figures."New value: +"Human-readable English sentence summarizing the projection. Always ends with 'Payouts follow a fixed milestone schedule; payout amounts are not guaranteed. See full disclosures at <url>.' — display the URL verbatim alongside any figures."

Previous value: -"Human-readable English sentence summarizing the projection. Always ends with 'Payouts are not guarantees. See full disclosures at <url>.' — display the URL verbatim alongside any figures."New value: +"Human-readable English sentence summarizing the projection. Always ends with 'Payouts follow a fixed milestone schedule; payout amounts are not guaranteed. See full disclosures at <url>.' — display the URL verbatim alongside any figures."

Previous value: -25New value: +18

Previous value: -25New value: +18

Previous value: -"Current age (default 40). Accepts 0–75; an age under 18 is modeled on the age-18 basis (growth understated), flagged via the response's `age_modeling` block"New value: +"Current age (default 40)"

Previous value: -0New value: +18

Removed value: -{
-  "description": "Disclosure (single-sourced from advisor_functions) that the projection is modeled on the age-18 basis and the shown growth is understated for a younger investor. Present only when clamped; display alongside any figures.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Present + true ONLY when the supplied current_age was below the age-18 matrix floor; the projection was then modeled on the age-18 basis.",
-  "type": "boolean"
-}

Removed value: -{
-  "description": "The age used for the matrix lookup (18) when current_age was younger. Present only when clamped.",
-  "type": "integer"
-}

Previous value: -"Investor's current age (default 40). Accepts 0–75; an age under 18 is modeled on the age-18 basis (the projection matrix has no rows below 18), so projected growth is understated for a sub-18 investor — the response flags this in its `age_modeling` block. Max 75 (max enrollment age)"New value: +"Investor's current age (default 40). Min 18 (the projection matrix floor); max 75 (max enrollment age)"

Previous value: -0New value: +18

Removed value: -{
-  "description": "Disclosure (single-sourced from advisor_functions) that the projection is modeled on the age-18 basis and the shown growth is understated for a younger investor. Present only when clamped; display alongside any figures.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Present + true ONLY when the supplied current_age was below the age-18 matrix floor; the projection was then modeled on the age-18 basis.",
-  "type": "boolean"
-}

Removed value: -{
-  "description": "The age used for the matrix lookup (18) when current_age was younger. Present only when clamped.",
-  "type": "integer"
-}

Previous value: -"Investor's current age (default 40). Accepts 0–75; an age under 18 is modeled on the age-18 basis (the projection matrix has no rows below 18), so projected growth is understated for a sub-18 investor — the response flags this in its `age_modeling` block. Max 75 (max enrollment age)"New value: +"Investor's current age (default 40). Min 18 (the projection matrix floor); max 75 (max enrollment age)"

Previous value: -0New value: +18

Removed value: -{
-  "description": "Disclosure (single-sourced from advisor_functions) that the projection is modeled on the age-18 basis and the shown growth is understated for a younger investor. Present only when clamped; display alongside any figures.",
-  "type": "string"
-}

Removed value: -{
-  "description": "Present + true ONLY when the supplied current_age was below the age-18 matrix floor; the projection was then modeled on the age-18 basis.",
-  "type": "boolean"
-}

Removed value: -{
-  "description": "The age used for the matrix lookup (18) when current_age was younger. Present only when clamped.",
-  "type": "integer"
-}