Hive Intelligence MCP
Evidence-backed crypto due diligence with sources, freshness, and a runtime receipt on every call.
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
~7,945token 数（工具定义）
~12.1 KB典型响应大小
对注意力有显著影响（占 128k 上下文窗口的 6.21%）
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": {
"mcp": {
"command": "npx",
"args": [
"hive-intelligence"
]
}
}
}可运行的软件包
npmhive-intelligence
1.9.0stdio远程端点
https://mcp.hiveintelligence.xyz/mcpstreamable-http它能做什么
工具清单
工具（8）
🟢 只读🟡 写入🔴 删除⚪ 未知
🟢get_token_price(token, chain, address, vs_currency, response_format)
Current price for one token. Pass token as a CoinGecko id or symbol ("bitcoin", "BTC"), or pass chain + address for a specific contract. Returns price, market cap, and 24h change with provider provenance.
输入模式
{
"type": "object",
"properties": {
"token": {
"description": "CoinGecko coin id or ticker symbol, for example \"bitcoin\" or \"BTC\". Omit when passing chain + address.",
"type": "string",
"minLength": 1,
"maxLength": 100
},
"chain": {
"description": "Chain name for the contract-address variant, for example \"eth\", \"base\", \"bsc\", or \"solana\".",
"type": "string",
"maxLength": 40
},
"address": {
"description": "Token contract address; requires chain.",
"type": "string",
"maxLength": 100
},
"vs_currency": {
"description": "Quote currency. Default \"usd\".",
"type": "string",
"maxLength": 20
},
"response_format": {
"description": "concise (default) returns the trimmed essentials; detailed returns the full provider payload. json and markdown, the platform-wide values every other tool advertises, are accepted and treated as concise.",
"type": "string",
"enum": [
"concise",
"detailed",
"json",
"markdown"
]
}
},
"additionalProperties": true
}输出模式
{
"type": "object",
"properties": {
"source_receipt": {
"type": "object",
"additionalProperties": true,
"required": [
"provider",
"runtime_status",
"fetched_at"
],
"properties": {
"provider": {
"type": "string"
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
}
}
},
"token": {
"type": "string"
},
"vs_currency": {
"type": "string"
},
"price": {
"type": "number"
},
"market_cap": {
"type": "number"
},
"change_24h_percent": {
"type": "number"
}
},
"required": [
"source_receipt"
],
"additionalProperties": true
}🟢check_token_safety(address, chain, response_format)
Token security screen for a contract address: honeypot, taxes, ownership, and related GoPlus risk datapoints. EVM chains use chain + address; Solana addresses are detected automatically.
输入模式
{
"type": "object",
"properties": {
"address": {
"type": "string",
"minLength": 1,
"maxLength": 100,
"description": "Required. Token contract address (0x… for EVM, base58 for Solana)."
},
"chain": {
"description": "Chain name or EVM chain id, for example \"eth\", \"bsc\", \"8453\", or \"solana\". Defaults to eth for 0x addresses; Solana is auto-detected.",
"type": "string",
"maxLength": 40
},
"response_format": {
"description": "concise (default) returns the trimmed essentials; detailed returns the full provider payload. json and markdown, the platform-wide values every other tool advertises, are accepted and treated as concise.",
"type": "string",
"enum": [
"concise",
"detailed",
"json",
"markdown"
]
}
},
"additionalProperties": true
}输出模式
{
"type": "object",
"properties": {
"source_receipt": {
"type": "object",
"additionalProperties": true,
"required": [
"provider",
"runtime_status",
"fetched_at"
],
"properties": {
"provider": {
"type": "string"
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
}
}
},
"address": {
"type": "string"
},
"chain": {
"type": "string"
},
"risk_flags": {
"type": "object",
"additionalProperties": true,
"description": "Raw provider risk datapoints; '1' means flagged. A missing key means the provider did not report that signal, not that it passed."
},
"holder_count": {},
"token_name": {
"type": "string"
},
"note": {
"type": "string"
}
},
"required": [
"source_receipt"
],
"additionalProperties": true
}🟢get_wallet_portfolio(address, chain, response_format)
Token holdings for a wallet with balances and USD values where available. EVM chains route through Alchemy; Solana wallets are detected automatically and route through Helius.
输入模式
{
"type": "object",
"properties": {
"address": {
"type": "string",
"minLength": 1,
"maxLength": 100,
"description": "Required. Wallet address (0x… for EVM, base58 for Solana)."
},
"chain": {
"description": "Chain name, for example \"eth\", \"base\", \"arbitrum\", or \"solana\". Defaults to eth for 0x addresses; Solana is auto-detected.",
"type": "string",
"maxLength": 40
},
"response_format": {
"description": "concise (default) returns the trimmed essentials; detailed returns the full provider payload. json and markdown, the platform-wide values every other tool advertises, are accepted and treated as concise.",
"type": "string",
"enum": [
"concise",
"detailed",
"json",
"markdown"
]
}
},
"additionalProperties": true
}输出模式
{
"type": "object",
"properties": {
"source_receipt": {
"type": "object",
"additionalProperties": true,
"required": [
"provider",
"runtime_status",
"fetched_at"
],
"properties": {
"provider": {
"type": "string"
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
}
}
},
"address": {
"type": "string"
},
"chain": {
"type": "string"
},
"token_count": {
"type": "number",
"description": "Total holdings found, which may exceed tokens.length."
},
"tokens": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
},
"description": "Holdings; truncated in concise mode when note is present."
},
"note": {
"type": "string"
}
},
"required": [
"source_receipt"
],
"additionalProperties": true
}🟢search_tools(query, provider, category, limit, cursor, ...)
Search canonical task toolsets, compact user-intent routes, and the full Hive coverage catalog. Select a matching route first, follow its ordered primary calls and conditional fallbacks, then inspect each exact schema. Execute reads with invoke_api_endpoint and explicitly approved Hive state changes with invoke_stateful_endpoint.
输入模式
{
"type": "object",
"properties": {
"query": {
"description": "Search text matched against tool name, title, category, provider, and summary.",
"type": "string",
"maxLength": 512
},
"provider": {
"description": "Optional provider filter such as CoinGecko, Moralis, Tenderly, Helius, GoPlus, CCXT, Alchemy, DeFiLlama, Hyperliquid, or Open Data Fetch.",
"type": "string",
"maxLength": 128
},
"category": {
"description": "Optional category filter, for example Portfolio & Wallet.",
"type": "string",
"maxLength": 128
},
"limit": {
"description": "Maximum tools to return. Default 20.",
"type": "integer",
"minimum": 1,
"maximum": 50
},
"cursor": {
"description": "Pagination cursor from the previous response.",
"type": "string"
},
"toolset_id": {
"description": "Optional exact canonical task toolset id. Use this after an initial compact result to inspect one workflow.",
"type": "string",
"maxLength": 100
},
"route_id": {
"description": "Optional exact task-shaped route id. Use the matched route's ordered calls and conditional fallbacks instead of choosing from the coverage catalog.",
"type": "string",
"maxLength": 100
},
"evidence_phase": {
"description": "Optional exact workflow evidence phase. Supported values: activity, alerts, approvals, assets, authenticity, balances, blocks, bridges, derivatives, discovery, enrichment, evidence, fees, funding, gas, history, holders, identity, infrastructure, liquidity, logs, market, markets, memory, monitoring, ohlcv, owners, positions, reports, risk, routing, sales, schema, simulation, stablecoins, trades, transactions, tvl, yields.",
"type": "string",
"maxLength": 100
},
"detail": {
"description": "Toolset detail level. Defaults to compact. Use full only when the task output schema is needed.",
"type": "string",
"enum": [
"compact",
"standard",
"full"
]
},
"toolset_limit": {
"description": "Maximum matching task toolsets to return. Default 3, max 5.",
"type": "integer",
"minimum": 1,
"maximum": 5
},
"toolset_cursor": {
"description": "Pagination cursor for task toolset matches.",
"type": "string"
},
"response_format": {
"type": "string",
"enum": [
"json",
"markdown"
],
"description": "Output format: 'json' for structured data (default) or 'markdown' for human-readable text."
}
},
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"_hive": {
"type": "object",
"additionalProperties": false,
"required": [
"receipt_id",
"receipt_version",
"server_version",
"build_sha",
"digest_algorithm",
"input_digest",
"result_digest",
"provider",
"tool",
"category",
"runtime_status",
"fetched_at",
"observed_at",
"cache_age_ms",
"duration_ms",
"cache_status",
"source",
"origin_source",
"truncated",
"warnings"
],
"properties": {
"receipt_id": {
"type": "string",
"format": "uuid",
"description": "Unique runtime receipt for this Hive tool call. Include it when reporting or debugging the evidence."
},
"receipt_version": {
"type": "string",
"const": "1.0"
},
"doc_url": {
"type": "string",
"description": "Docs anchor for the error code (errors only)."
},
"credit_cost": {
"type": "integer",
"enum": [
0,
1
],
"description": "Static credit price of this tool: 0 for discovery tools and free writes, 1 for everything else. Present when the calling lane bills."
},
"credits_used": {
"type": "integer",
"minimum": 0,
"description": "Credits actually debited for this call: 0 when validation or a gate stopped it, credit_cost once the provider was reached."
},
"credits_remaining": {
"anyOf": [
{
"type": "integer",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Balance after this call: keyed credits, or keyless calls left today. null for unlimited plans; absent when the lane has no notion."
},
"server_version": {
"type": "string"
},
"build_sha": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"digest_algorithm": {
"type": "string",
"const": "sha256"
},
"input_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of canonical JSON arguments. It is not a server signature."
},
"result_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of the canonical returned payload after Hive result normalization and before the top-level _hive receipt is attached."
},
"provider": {
"type": "string"
},
"tool": {
"type": "string"
},
"category": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
},
"observed_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"description": "When Hive first observed this response. On a cache hit this is the original cache-population time; it is not necessarily the upstream datum's event time."
},
"cache_age_ms": {
"anyOf": [
{
"type": "number",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Milliseconds since Hive first observed the response. Zero means newly retrieved from the provider, not that the provider's underlying datum is current."
},
"duration_ms": {
"type": "number"
},
"cache_status": {
"type": "string",
"enum": [
"miss",
"hit",
"bypass",
"unknown"
]
},
"source": {
"type": "string",
"enum": [
"live",
"cached",
"fallback",
"unavailable"
]
},
"origin_source": {
"anyOf": [
{
"type": "string",
"enum": [
"live",
"fallback"
]
},
{
"type": "null"
}
],
"description": "Original upstream tier. Preserved as fallback when a fallback-tier response is later served from cache."
},
"truncated": {
"type": "boolean"
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
},
"fallback_used": {
"type": "boolean",
"description": "Present and true when a task-route fallback tool produced this response after the primary returned failing/missing_key. provider/tool name the actual executed source."
},
"sanitized": {
"type": "boolean",
"description": "Present and true when instruction-shaped provider text was neutralized before reaching agent context."
},
"cause": {
"type": "string",
"description": "Errors only: what actually happened, written for the model."
},
"next_action": {
"type": "string",
"description": "Errors only: the one useful next step. Follow it instead of retrying the same call unchanged."
}
}
},
"query": {
"type": "string"
},
"route_id": {
"type": "string"
},
"provider": {
"type": "string"
},
"category": {
"type": "string"
},
"tools": {
"type": "array",
"items": {
"type": "object",
"properties": {
"name": {
"type": "string"
},
"title": {
"type": "string"
},
"category": {
"type": "string"
},
"provider": {
"type": "string"
},
"material": {
"type": "boolean",
"description": "True when the call costs one credit (or one keyless call); false for free discovery tools and free writes."
},
"credit_cost": {
"type": "integer",
"enum": [
0,
1
]
},
"summary": {
"type": "string"
},
"operation": {
"type": "string",
"enum": [
"read",
"write"
]
},
"callableVia": {
"type": "string",
"enum": [
"invoke_api_endpoint",
"invoke_stateful_endpoint"
]
},
"readOnly": {
"type": "boolean"
},
"hasOutputSchema": {
"type": "boolean"
},
"inputRequired": {
"type": "array",
"items": {
"type": "string"
}
},
"availability": {
"type": "string",
"enum": [
"ok",
"degraded",
"missing_key",
"failing"
],
"description": "Live availability of this tool's provider from the cached health snapshot. missing_key/failing tools are ranked last and return runtime_status, not data."
},
"availability_note": {
"type": "string"
}
}
}
},
"toolsets": {
"type": "array",
"items": {
"type": "object",
"properties": {
"id": {
"type": "string"
},
"label": {
"type": "string"
},
"kind": {
"type": "string"
},
"purpose": {
"type": "string"
},
"outcome": {
"type": "string"
},
"requiredIdentifiers": {
"type": "array",
"items": {
"type": "string"
}
},
"routes": {
"type": "array",
"items": {
"type": "object",
"additionalProperties": true
}
},
"coverageCatalog": {
"type": "object",
"additionalProperties": true
},
"providers": {
"type": "array",
"items": {
"type": "string"
}
},
"tags": {
"type": "array",
"items": {
"type": "string"
}
},
"resource": {
"type": "string"
},
"runtimeStatuses": {
"type": "array",
"items": {
"type": "string"
}
},
"executionPolicy": {
"type": "object",
"additionalProperties": true
},
"outputSchema": {
"type": "object",
"additionalProperties": true
},
"providerAvailability": {
"type": "object",
"additionalProperties": true,
"description": "Present only when a provider in this toolset is not ok: provider name -> degraded | missing_key | failing."
},
"availability_note": {
"type": "string"
}
}
}
},
"toolset_detail": {
"type": "string",
"enum": [
"compact",
"standard",
"full"
]
},
"toolset_pagination": {
"type": "object",
"properties": {
"returned": {
"type": "number"
},
"total": {
"type": "number"
},
"limit": {
"type": "number"
},
"cursor": {
"type": "string"
},
"has_more": {
"type": "boolean"
},
"next_cursor": {
"type": "string"
}
}
},
"pagination": {
"type": "object",
"properties": {
"returned": {
"type": "number"
},
"total": {
"type": "number"
},
"limit": {
"type": "number"
},
"cursor": {
"type": "string"
},
"has_more": {
"type": "boolean"
},
"next_cursor": {
"type": "string"
}
}
}
},
"required": [
"_hive"
]
}🟢get_api_endpoint_schema(endpoint, endpoint_name, response_format)
Returns the full input schema, operation type, and correct root invoker for any Hive tool by name. Tool names come from search_tools; inspect this contract before execution.
输入模式
{
"type": "object",
"properties": {
"endpoint": {
"description": "The name of the endpoint to get the schema for.",
"type": "string"
},
"endpoint_name": {
"description": "Alias for 'endpoint', accepted for symmetry with invoke_api_endpoint. Provide one of the two.",
"type": "string"
},
"response_format": {
"type": "string",
"enum": [
"json",
"markdown"
],
"description": "Output format: 'json' for structured data (default) or 'markdown' for human-readable text."
}
},
"additionalProperties": true
}输出模式
{
"type": "object",
"properties": {
"_hive": {
"type": "object",
"additionalProperties": false,
"required": [
"receipt_id",
"receipt_version",
"server_version",
"build_sha",
"digest_algorithm",
"input_digest",
"result_digest",
"provider",
"tool",
"category",
"runtime_status",
"fetched_at",
"observed_at",
"cache_age_ms",
"duration_ms",
"cache_status",
"source",
"origin_source",
"truncated",
"warnings"
],
"properties": {
"receipt_id": {
"type": "string",
"format": "uuid",
"description": "Unique runtime receipt for this Hive tool call. Include it when reporting or debugging the evidence."
},
"receipt_version": {
"type": "string",
"const": "1.0"
},
"doc_url": {
"type": "string",
"description": "Docs anchor for the error code (errors only)."
},
"credit_cost": {
"type": "integer",
"enum": [
0,
1
],
"description": "Static credit price of this tool: 0 for discovery tools and free writes, 1 for everything else. Present when the calling lane bills."
},
"credits_used": {
"type": "integer",
"minimum": 0,
"description": "Credits actually debited for this call: 0 when validation or a gate stopped it, credit_cost once the provider was reached."
},
"credits_remaining": {
"anyOf": [
{
"type": "integer",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Balance after this call: keyed credits, or keyless calls left today. null for unlimited plans; absent when the lane has no notion."
},
"server_version": {
"type": "string"
},
"build_sha": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"digest_algorithm": {
"type": "string",
"const": "sha256"
},
"input_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of canonical JSON arguments. It is not a server signature."
},
"result_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of the canonical returned payload after Hive result normalization and before the top-level _hive receipt is attached."
},
"provider": {
"type": "string"
},
"tool": {
"type": "string"
},
"category": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
},
"observed_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"description": "When Hive first observed this response. On a cache hit this is the original cache-population time; it is not necessarily the upstream datum's event time."
},
"cache_age_ms": {
"anyOf": [
{
"type": "number",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Milliseconds since Hive first observed the response. Zero means newly retrieved from the provider, not that the provider's underlying datum is current."
},
"duration_ms": {
"type": "number"
},
"cache_status": {
"type": "string",
"enum": [
"miss",
"hit",
"bypass",
"unknown"
]
},
"source": {
"type": "string",
"enum": [
"live",
"cached",
"fallback",
"unavailable"
]
},
"origin_source": {
"anyOf": [
{
"type": "string",
"enum": [
"live",
"fallback"
]
},
{
"type": "null"
}
],
"description": "Original upstream tier. Preserved as fallback when a fallback-tier response is later served from cache."
},
"truncated": {
"type": "boolean"
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
},
"fallback_used": {
"type": "boolean",
"description": "Present and true when a task-route fallback tool produced this response after the primary returned failing/missing_key. provider/tool name the actual executed source."
},
"sanitized": {
"type": "boolean",
"description": "Present and true when instruction-shaped provider text was neutralized before reaching agent context."
},
"cause": {
"type": "string",
"description": "Errors only: what actually happened, written for the model."
},
"next_action": {
"type": "string",
"description": "Errors only: the one useful next step. Follow it instead of retrying the same call unchanged."
}
}
},
"name": {
"type": "string"
},
"title": {
"type": "string"
},
"description": {
"type": "string"
},
"category": {
"type": "string"
},
"provider": {
"type": "string"
},
"inputSchema": {
"type": "object",
"additionalProperties": true
},
"outputSchema": {
"type": "object",
"additionalProperties": true
},
"annotations": {
"type": "object",
"additionalProperties": true
},
"operation": {
"type": "string",
"enum": [
"read",
"write"
]
},
"toolsets": {
"type": "array",
"items": {
"type": "object",
"properties": {
"id": {
"type": "string"
},
"label": {
"type": "string"
},
"kind": {
"type": "string"
}
}
}
},
"runtimeStatuses": {
"type": "array",
"items": {
"type": "string"
}
},
"runtimeStatusSemantics": {
"type": "object",
"additionalProperties": true
},
"callPattern": {
"type": "string"
}
},
"required": [
"_hive"
],
"additionalProperties": true
}🟢invoke_api_endpoint(endpoint_name, endpoint, args, response_format)
Invokes only read-only Hive endpoints by name. First discover the tool, inspect get_api_endpoint_schema, and pass bounded schema-valid arguments. State-changing endpoints are rejected and must use invoke_stateful_endpoint after explicit user approval.
输入模式
{
"type": "object",
"properties": {
"endpoint_name": {
"description": "The name of the endpoint to invoke.",
"type": "string"
},
"endpoint": {
"description": "Alias for 'endpoint_name', accepted by both root invokers. Provide one of the two.",
"type": "string"
},
"args": {
"description": "The arguments to pass to the endpoint. This must match the schema returned by the `get_api_endpoint_schema` tool. You may also pass arguments as flat top-level keys alongside endpoint_name.",
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {}
},
"response_format": {
"type": "string",
"enum": [
"json",
"markdown"
],
"description": "Output format: 'json' for structured data (default) or 'markdown' for human-readable text."
}
},
"additionalProperties": {}
}输出模式
{
"type": "object",
"properties": {
"_hive": {
"type": "object",
"additionalProperties": false,
"required": [
"receipt_id",
"receipt_version",
"server_version",
"build_sha",
"digest_algorithm",
"input_digest",
"result_digest",
"provider",
"tool",
"category",
"runtime_status",
"fetched_at",
"observed_at",
"cache_age_ms",
"duration_ms",
"cache_status",
"source",
"origin_source",
"truncated",
"warnings"
],
"properties": {
"receipt_id": {
"type": "string",
"format": "uuid",
"description": "Unique runtime receipt for this Hive tool call. Include it when reporting or debugging the evidence."
},
"receipt_version": {
"type": "string",
"const": "1.0"
},
"doc_url": {
"type": "string",
"description": "Docs anchor for the error code (errors only)."
},
"credit_cost": {
"type": "integer",
"enum": [
0,
1
],
"description": "Static credit price of this tool: 0 for discovery tools and free writes, 1 for everything else. Present when the calling lane bills."
},
"credits_used": {
"type": "integer",
"minimum": 0,
"description": "Credits actually debited for this call: 0 when validation or a gate stopped it, credit_cost once the provider was reached."
},
"credits_remaining": {
"anyOf": [
{
"type": "integer",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Balance after this call: keyed credits, or keyless calls left today. null for unlimited plans; absent when the lane has no notion."
},
"server_version": {
"type": "string"
},
"build_sha": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"digest_algorithm": {
"type": "string",
"const": "sha256"
},
"input_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of canonical JSON arguments. It is not a server signature."
},
"result_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of the canonical returned payload after Hive result normalization and before the top-level _hive receipt is attached."
},
"provider": {
"type": "string"
},
"tool": {
"type": "string"
},
"category": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
},
"observed_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"description": "When Hive first observed this response. On a cache hit this is the original cache-population time; it is not necessarily the upstream datum's event time."
},
"cache_age_ms": {
"anyOf": [
{
"type": "number",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Milliseconds since Hive first observed the response. Zero means newly retrieved from the provider, not that the provider's underlying datum is current."
},
"duration_ms": {
"type": "number"
},
"cache_status": {
"type": "string",
"enum": [
"miss",
"hit",
"bypass",
"unknown"
]
},
"source": {
"type": "string",
"enum": [
"live",
"cached",
"fallback",
"unavailable"
]
},
"origin_source": {
"anyOf": [
{
"type": "string",
"enum": [
"live",
"fallback"
]
},
{
"type": "null"
}
],
"description": "Original upstream tier. Preserved as fallback when a fallback-tier response is later served from cache."
},
"truncated": {
"type": "boolean"
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
},
"fallback_used": {
"type": "boolean",
"description": "Present and true when a task-route fallback tool produced this response after the primary returned failing/missing_key. provider/tool name the actual executed source."
},
"sanitized": {
"type": "boolean",
"description": "Present and true when instruction-shaped provider text was neutralized before reaching agent context."
},
"cause": {
"type": "string",
"description": "Errors only: what actually happened, written for the model."
},
"next_action": {
"type": "string",
"description": "Errors only: the one useful next step. Follow it instead of retrying the same call unchanged."
}
}
},
"items": {
"type": "array",
"items": {},
"description": "Array result when a provider returns a top-level list."
},
"data": {
"description": "Provider data payload when the response uses a data wrapper."
},
"result": {
"description": "Provider result payload when the response uses a result wrapper."
},
"error": {
"type": "string"
},
"message": {
"type": "string"
},
"status": {
"description": "Provider-reported status. Type varies by provider (string, boolean, or number), so no type constraint is applied."
},
"_metadata": {
"type": "object",
"additionalProperties": true
},
"_pagination": {
"type": "object",
"additionalProperties": true,
"properties": {
"returned": {
"type": "number"
},
"total_available": {
"type": "number"
},
"has_more": {
"type": "boolean"
}
}
},
"_provider": {
"type": "string"
},
"_degraded": {
"type": "string"
},
"_truncated": {
"type": "boolean"
},
"_type": {
"type": "string"
}
},
"required": [
"_hive"
],
"additionalProperties": true
}🔴invoke_stateful_endpoint(endpoint_name, endpoint, args, response_format)
Invokes a Hive-native state-changing endpoint after explicit user approval. This router can create, update, archive, remember, forget, acknowledge, or resolve durable Hive state, so clients must not auto-approve it. Inspect get_api_endpoint_schema first and verify the saved result after execution.
输入模式
{
"type": "object",
"properties": {
"endpoint_name": {
"description": "The name of the endpoint to invoke.",
"type": "string"
},
"endpoint": {
"description": "Alias for 'endpoint_name', accepted by both root invokers. Provide one of the two.",
"type": "string"
},
"args": {
"description": "The arguments to pass to the endpoint. This must match the schema returned by the `get_api_endpoint_schema` tool. You may also pass arguments as flat top-level keys alongside endpoint_name.",
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {}
},
"response_format": {
"type": "string",
"enum": [
"json",
"markdown"
],
"description": "Output format: 'json' for structured data (default) or 'markdown' for human-readable text."
}
},
"additionalProperties": {}
}输出模式
{
"type": "object",
"properties": {
"_hive": {
"type": "object",
"additionalProperties": false,
"required": [
"receipt_id",
"receipt_version",
"server_version",
"build_sha",
"digest_algorithm",
"input_digest",
"result_digest",
"provider",
"tool",
"category",
"runtime_status",
"fetched_at",
"observed_at",
"cache_age_ms",
"duration_ms",
"cache_status",
"source",
"origin_source",
"truncated",
"warnings"
],
"properties": {
"receipt_id": {
"type": "string",
"format": "uuid",
"description": "Unique runtime receipt for this Hive tool call. Include it when reporting or debugging the evidence."
},
"receipt_version": {
"type": "string",
"const": "1.0"
},
"doc_url": {
"type": "string",
"description": "Docs anchor for the error code (errors only)."
},
"credit_cost": {
"type": "integer",
"enum": [
0,
1
],
"description": "Static credit price of this tool: 0 for discovery tools and free writes, 1 for everything else. Present when the calling lane bills."
},
"credits_used": {
"type": "integer",
"minimum": 0,
"description": "Credits actually debited for this call: 0 when validation or a gate stopped it, credit_cost once the provider was reached."
},
"credits_remaining": {
"anyOf": [
{
"type": "integer",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Balance after this call: keyed credits, or keyless calls left today. null for unlimited plans; absent when the lane has no notion."
},
"server_version": {
"type": "string"
},
"build_sha": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"digest_algorithm": {
"type": "string",
"const": "sha256"
},
"input_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of canonical JSON arguments. It is not a server signature."
},
"result_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of the canonical returned payload after Hive result normalization and before the top-level _hive receipt is attached."
},
"provider": {
"type": "string"
},
"tool": {
"type": "string"
},
"category": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
},
"observed_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"description": "When Hive first observed this response. On a cache hit this is the original cache-population time; it is not necessarily the upstream datum's event time."
},
"cache_age_ms": {
"anyOf": [
{
"type": "number",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Milliseconds since Hive first observed the response. Zero means newly retrieved from the provider, not that the provider's underlying datum is current."
},
"duration_ms": {
"type": "number"
},
"cache_status": {
"type": "string",
"enum": [
"miss",
"hit",
"bypass",
"unknown"
]
},
"source": {
"type": "string",
"enum": [
"live",
"cached",
"fallback",
"unavailable"
]
},
"origin_source": {
"anyOf": [
{
"type": "string",
"enum": [
"live",
"fallback"
]
},
{
"type": "null"
}
],
"description": "Original upstream tier. Preserved as fallback when a fallback-tier response is later served from cache."
},
"truncated": {
"type": "boolean"
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
},
"fallback_used": {
"type": "boolean",
"description": "Present and true when a task-route fallback tool produced this response after the primary returned failing/missing_key. provider/tool name the actual executed source."
},
"sanitized": {
"type": "boolean",
"description": "Present and true when instruction-shaped provider text was neutralized before reaching agent context."
},
"cause": {
"type": "string",
"description": "Errors only: what actually happened, written for the model."
},
"next_action": {
"type": "string",
"description": "Errors only: the one useful next step. Follow it instead of retrying the same call unchanged."
}
}
},
"items": {
"type": "array",
"items": {},
"description": "Array result when a provider returns a top-level list."
},
"data": {
"description": "Provider data payload when the response uses a data wrapper."
},
"result": {
"description": "Provider result payload when the response uses a result wrapper."
},
"error": {
"type": "string"
},
"message": {
"type": "string"
},
"status": {
"description": "Provider-reported status. Type varies by provider (string, boolean, or number), so no type constraint is applied."
},
"_metadata": {
"type": "object",
"additionalProperties": true
},
"_pagination": {
"type": "object",
"additionalProperties": true,
"properties": {
"returned": {
"type": "number"
},
"total_available": {
"type": "number"
},
"has_more": {
"type": "boolean"
}
}
},
"_provider": {
"type": "string"
},
"_degraded": {
"type": "string"
},
"_truncated": {
"type": "boolean"
},
"_type": {
"type": "string"
}
},
"required": [
"_hive"
],
"additionalProperties": true
}🟢validate_task_result(task_toolset_id, result, response_format)
Validates a proposed workflow answer against Hive's typed task-output and evidence-receipt contract before it is shown to the user. It checks the selected route id, ordered route-only calls, conditional fallback use, the four-call route budget, claim-to-receipt citations, canonical phase coverage, and internal receipt consistency. It cannot authenticate invented receipts or turn SHA-256 self-checks into signatures, so every call entry must be copied exactly from the server-minted _hive block returned by that execution.
输入模式
{
"type": "object",
"properties": {
"task_toolset_id": {
"type": "string",
"description": "Canonical task toolset id returned by search_tools."
},
"result": {
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {},
"description": "The proposed task result envelope. Each receipt.calls entry must be copied exactly from a material call's _hive object."
},
"response_format": {
"type": "string",
"enum": [
"json",
"markdown"
],
"description": "Output format: 'json' for structured data (default) or 'markdown' for human-readable text."
}
},
"required": [
"task_toolset_id",
"result"
],
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"_hive": {
"type": "object",
"additionalProperties": false,
"required": [
"receipt_id",
"receipt_version",
"server_version",
"build_sha",
"digest_algorithm",
"input_digest",
"result_digest",
"provider",
"tool",
"category",
"runtime_status",
"fetched_at",
"observed_at",
"cache_age_ms",
"duration_ms",
"cache_status",
"source",
"origin_source",
"truncated",
"warnings"
],
"properties": {
"receipt_id": {
"type": "string",
"format": "uuid",
"description": "Unique runtime receipt for this Hive tool call. Include it when reporting or debugging the evidence."
},
"receipt_version": {
"type": "string",
"const": "1.0"
},
"doc_url": {
"type": "string",
"description": "Docs anchor for the error code (errors only)."
},
"credit_cost": {
"type": "integer",
"enum": [
0,
1
],
"description": "Static credit price of this tool: 0 for discovery tools and free writes, 1 for everything else. Present when the calling lane bills."
},
"credits_used": {
"type": "integer",
"minimum": 0,
"description": "Credits actually debited for this call: 0 when validation or a gate stopped it, credit_cost once the provider was reached."
},
"credits_remaining": {
"anyOf": [
{
"type": "integer",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Balance after this call: keyed credits, or keyless calls left today. null for unlimited plans; absent when the lane has no notion."
},
"server_version": {
"type": "string"
},
"build_sha": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"digest_algorithm": {
"type": "string",
"const": "sha256"
},
"input_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of canonical JSON arguments. It is not a server signature."
},
"result_digest": {
"type": "string",
"pattern": "^[a-f0-9]{64}$",
"description": "SHA-256 self-check of the canonical returned payload after Hive result normalization and before the top-level _hive receipt is attached."
},
"provider": {
"type": "string"
},
"tool": {
"type": "string"
},
"category": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"runtime_status": {
"type": "string",
"enum": [
"ok",
"invalid_input",
"missing_key",
"plan_required",
"rate_limited",
"degraded",
"failing"
]
},
"fetched_at": {
"type": "string",
"format": "date-time"
},
"observed_at": {
"anyOf": [
{
"type": "string",
"format": "date-time"
},
{
"type": "null"
}
],
"description": "When Hive first observed this response. On a cache hit this is the original cache-population time; it is not necessarily the upstream datum's event time."
},
"cache_age_ms": {
"anyOf": [
{
"type": "number",
"minimum": 0
},
{
"type": "null"
}
],
"description": "Milliseconds since Hive first observed the response. Zero means newly retrieved from the provider, not that the provider's underlying datum is current."
},
"duration_ms": {
"type": "number"
},
"cache_status": {
"type": "string",
"enum": [
"miss",
"hit",
"bypass",
"unknown"
]
},
"source": {
"type": "string",
"enum": [
"live",
"cached",
"fallback",
"unavailable"
]
},
"origin_source": {
"anyOf": [
{
"type": "string",
"enum": [
"live",
"fallback"
]
},
{
"type": "null"
}
],
"description": "Original upstream tier. Preserved as fallback when a fallback-tier response is later served from cache."
},
"truncated": {
"type": "boolean"
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
},
"fallback_used": {
"type": "boolean",
"description": "Present and true when a task-route fallback tool produced this response after the primary returned failing/missing_key. provider/tool name the actual executed source."
},
"sanitized": {
"type": "boolean",
"description": "Present and true when instruction-shaped provider text was neutralized before reaching agent context."
},
"cause": {
"type": "string",
"description": "Errors only: what actually happened, written for the model."
},
"next_action": {
"type": "string",
"description": "Errors only: the one useful next step. Follow it instead of retrying the same call unchanged."
}
}
},
"valid": {
"type": "boolean"
},
"task_toolset_id": {
"type": "string"
},
"route_id": {
"type": "string"
},
"call_count": {
"type": "number"
},
"receipt_ids": {
"type": "array",
"items": {
"type": "string"
}
},
"errors": {
"type": "array",
"items": {
"type": "object"
}
},
"warnings": {
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"valid",
"task_toolset_id",
"route_id",
"call_count",
"receipt_ids",
"errors",
"warnings",
"_hive"
],
"additionalProperties": false
}社区
证据
最近观测
已验证未记录版本8 个工具
已验证未记录版本8 个工具