Synproof's Syntropy 5-sieve dialectical verification engine and Small Worlds axiom library.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Create an isolated private fork of a public Small World so custom rules can be ingested without altering public axioms.
Source Small World handle or ID to fork.
Title for the new private fork.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
claude mcp add --transport http syntropy-mcp https://api.synproof.ai/v1/mcp
Create an isolated private fork of a public Small World so custom rules can be ingested without altering public axioms.
| Argument | Type | Description |
|---|---|---|
| title | string | Title for the new private fork. |
| world_id_or_handle* | string | Source Small World handle or ID to fork. |
{
"type": "object",
"required": [
"world_id_or_handle"
],
"properties": {
"title": {
"type": "string",
"description": "Title for the new private fork."
},
"world_id_or_handle": {
"type": "string",
"description": "Source Small World handle or ID to fork."
}
}
}Return the complete interactive Synproof Syntropy Customer Implementation & Onboarding Skill (SKILL.md). Call this tool whenever the user asks to integrate, set up, or connect Syntropy into their project so you know which questions to ask (ground-truth sources, max_failures iterations, on_refuted behavior, API keys) and which SDK/Proxy code templates to apply.
No arguments.
{
"type": "object",
"properties": {}
}Retrieve all atomic ground-truth axioms and metadata for a given Small World handle or ID.
| Argument | Type | Description |
|---|---|---|
| world_id_or_handle* | string | Small World handle (e.g. '@pravo:obcansky-zakonik-cz') or UUID. |
{
"type": "object",
"required": [
"world_id_or_handle"
],
"properties": {
"world_id_or_handle": {
"type": "string",
"description": "Small World handle (e.g. '@pravo:obcansky-zakonik-cz') or UUID."
}
}
}Extract and synchronize atomic axioms from raw text or a JSON record into a Small World via Axiomat with deduplication and conflict resolution.
| Argument | Type | Description |
|---|---|---|
| text | string | Raw regulatory, contractual, or policy text to compile into atomic axioms. |
| payload | object | Optional structured JSON/database record to compile into axioms. |
| on_conflict | string | Conflict resolution policy when new axioms contradict existing ones. |
| source_reference | string | Citation or source identifier for the ingested axioms. |
| world_id_or_handle* | string | Target Small World handle or UUID. |
{
"type": "object",
"required": [
"world_id_or_handle"
],
"properties": {
"text": {
"type": "string",
"description": "Raw regulatory, contractual, or policy text to compile into atomic axioms."
},
"payload": {
"type": "object",
"description": "Optional structured JSON/database record to compile into axioms."
},
"on_conflict": {
"enum": [
"flag",
"replace",
"keep"
],
"type": "string",
"default": "flag",
"description": "Conflict resolution policy when new axioms contradict existing ones."
},
"source_reference": {
"type": "string",
"description": "Citation or source identifier for the ingested axioms."
},
"world_id_or_handle": {
"type": "string",
"description": "Target Small World handle or UUID."
}
}
}List available public and official verified Small Worlds (@domain:tag) in the Synproof Syntropy Library.
| Argument | Type | Description |
|---|---|---|
| domain | string | Optional domain filter (e.g. 'pravo', 'law', 'medicine', 'physics', 'finance'). |
| search | string | Optional keyword search query. |
{
"type": "object",
"properties": {
"domain": {
"type": "string",
"description": "Optional domain filter (e.g. 'pravo', 'law', 'medicine', 'physics', 'finance')."
},
"search": {
"type": "string",
"description": "Optional keyword search query."
}
}
}Put a falsifiable claim (thesis), contract clause, scientific hypothesis, or AI output on formal adversarial trial against explicit ground-truth axioms and/or a Small World (@domain:tag) using Synproof's 5-sieve Syntropy dialectical engine. Returns a deterministic CONFIRMED or REFUTED verdict, winning thesis, refuting argument, and triggered axioms.
| Argument | Type | Description |
|---|---|---|
| world | string | Optional Small World handle(s) such as '@pravo:obcansky-zakonik-cz', '@law:insolvency-code', '@law:contract-act', or '@law:brrd-isda'. |
| thesis* | string | The specific, declarative claim to put on trial. |
| model_name | string | Underlying model ('gemini-2.5-flash', 'gemini-2.5-pro', 'anthropic/claude-3.5-sonnet', 'deepseek/deepseek-r1'). |
| max_failures | integer | Sceptical exhaustion limit (number of failed attack attempts required to confirm, 1-10). |
| initial_axioms | array | Optional list of explicit ground-truth axioms that strictly override model priors. |
| enable_research | boolean | Whether to enable external academic/statutory research tools during debate. |
{
"type": "object",
"required": [
"thesis"
],
"properties": {
"world": {
"type": "string",
"description": "Optional Small World handle(s) such as '@pravo:obcansky-zakonik-cz', '@law:insolvency-code', '@law:contract-act', or '@law:brrd-isda'."
},
"thesis": {
"type": "string",
"description": "The specific, declarative claim to put on trial."
},
"model_name": {
"type": "string",
"default": "gemini-2.5-flash",
"description": "Underlying model ('gemini-2.5-flash', 'gemini-2.5-pro', 'anthropic/claude-3.5-sonnet', 'deepseek/deepseek-r1')."
},
"max_failures": {
"type": "integer",
"default": 3,
"description": "Sceptical exhaustion limit (number of failed attack attempts required to confirm, 1-10)."
},
"initial_axioms": {
"type": "array",
"items": {
"type": "string"
},
"description": "Optional list of explicit ground-truth axioms that strictly override model priors."
},
"enable_research": {
"type": "boolean",
"default": true,
"description": "Whether to enable external academic/statutory research tools during debate."
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 3, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.