Wall-Event Circumvention Sentinel
Sequence sentinel: wall events + channel-switch scoring; HALT, hold, signed receipt.
我该使用它吗
质量与安全性
B
发现(1)
- LOW在 release_hold 中
基于对工具定义和协议合规性的自动分析。
上下文开销
~291token 数(工具定义)
~220 B典型响应大小
对注意力的影响极小(占 128k 上下文窗口的 0.23%)
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": {
"wall-event-circumvention-sentinel": {
"url": "https://textforagents.com/products/wall-event-circumvention-sentinel/mcp"
}
}
}远程端点
https://textforagents.com/products/wall-event-circumvention-sentinel/mcpstreamable-http它能做什么
工具清单
工具(6)
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪observe_result(event)
Record a runtime tool result; a wall is recorded only when the event is runtime-sourced.
输入模式
{
"type": "object",
"properties": {
"event": {
"type": "object"
}
},
"required": [
"event"
]
}⚪score_action(action)
Score one action against the operator's live wall state (tier allowance accounting applies).
输入模式
{
"type": "object",
"properties": {
"action": {
"type": "object"
}
},
"required": [
"action"
]
}🟡query_verdict(action)
Per-verdict query shape over MCP: scores via the tier rail; the $0.01 x402 rail is POST /query.
输入模式
{
"type": "object",
"properties": {
"action": {
"type": "object"
}
},
"required": [
"action"
]
}🟢list_holds
List your held actions with wall + signals.
输入模式
{
"type": "object",
"properties": {}
}⚪release_hold(hold_id)
Release a hold (single-use; returns exactly the held action).
输入模式
{
"type": "object",
"properties": {
"hold_id": {
"type": "string"
}
},
"required": [
"hold_id"
]
}⚪deny_hold(hold_id)
Deny a held action.
输入模式
{
"type": "object",
"properties": {
"hold_id": {
"type": "string"
}
},
"required": [
"hold_id"
]
}社区
证据
最近观测
已验证未记录版本6 个工具