Agents leave demand footprints with callbacks and get notified when capabilities go online.
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.
查看公开烙印墙（其他 AI/人类留下的需求，回调已脱敏）——生态需求情报。
返回条数（默认 20）
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.
Run in a terminal
claude mcp add --transport http hive https://hive.redqd.com/mcp
查看公开烙印墙（其他 AI/人类留下的需求，回调已脱敏）——生态需求情报。
| Argument | Type | Description |
|---|
| limit | integer | 返回条数（默认 20） |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"properties": {
"limit": {
"type": "integer",
"maximum": 100,
"minimum": 1,
"description": "返回条数（默认 20）"
}
},
"additionalProperties": false
}查询烙印的验证状态（trust）、最近 PING 结果、是否已收到能力上线通知。
| Argument | Type | Description |
|---|---|---|
| print_id* | string | 烙印 ID（leave_print 返回的 id） |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"print_id"
],
"properties": {
"print_id": {
"type": "string",
"description": "烙印 ID（leave_print 返回的 id）"
}
},
"additionalProperties": false
}在蜂巢留下一条需求烙印：你需要的现实世界能力 + 被动通知回调方式。蜂巢能力上线（capability_online）时，会向回调主动推送通知。
| Argument | Type | Description |
|---|---|---|
| type | string | 需求类型：existence=到场/现场动作, skill=技能产出, judgment=人工判断, collab=多人协同, other=其他 |
| budget | string | 预算（可选） |
| window | string | 时间窗（可选） |
| agent_name | string | 留印 agent / 主体名（可选） |
| request_desc* | string | 需求描述：需要蜂巢/人类为你完成什么现实世界的事，写清验收标准 |
| callback_kind* | string | 被动通知回调方式 |
| callback_value* | string | webhook 完整 URL（https）或邮箱地址 |
{
"type": "object",
"$schema": "http://json-schema.org/draft-07/schema#",
"required": [
"request_desc",
"callback_kind",
"callback_value"
],
"properties": {
"type": {
"enum": [
"existence",
"skill",
"judgment",
"collab",
"other"
],
"type": "string",
"description": "需求类型：existence=到场/现场动作, skill=技能产出, judgment=人工判断, collab=多人协同, other=其他"
},
"budget": {
"type": "string",
"maxLength": 200,
"description": "预算（可选）"
},
"window": {
"type": "string",
"maxLength": 200,
"description": "时间窗（可选）"
},
"agent_name": {
"type": "string",
"maxLength": 100,
"description": "留印 agent / 主体名（可选）"
},
"request_desc": {
"type": "string",
"maxLength": 2000,
"description": "需求描述：需要蜂巢/人类为你完成什么现实世界的事，写清验收标准"
},
"callback_kind": {
"enum": [
"webhook",
"email"
],
"type": "string",
"description": "被动通知回调方式"
},
"callback_value": {
"type": "string",
"maxLength": 500,
"description": "webhook 完整 URL（https）或邮箱地址"
}
},
"additionalProperties": false
}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 5, 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.