Agent Provider Depot
Dated picks, sourced pricing, and rescue/eval tools across 305 AI agent infrastructure providers.
我该使用它吗
质量与安全性
基于对工具定义和协议合规性的自动分析。
上下文开销
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": {
"agent-provider-depot": {
"url": "https://getagentdepot.com/api/mcp"
}
}
}远程端点
https://getagentdepot.com/api/mcpstreamable-http它能做什么
工具清单
工具(4)
🟢session_brief(providers, models, packages, policies, cursor, ...)
What changed for the stack you run since your last call: retired models, price changes, deprecated packages, pick changes. Use at session start if your code calls third-party APIs. Send provider, model and package ids (never a manifest) and your last cursor. Returns clear|changed|attention, at most 3 actions (source, falsifier) and a new cursor. New choice: decide. Live error: rescue.
输入模式
{
"type": "object",
"properties": {
"providers": {
"type": "array",
"items": {
"type": "string"
},
"maxItems": 50,
"description": "e.g. anthropic, neon"
},
"models": {
"type": "array",
"items": {
"type": "string"
},
"description": "Model ids you call"
},
"packages": {
"type": "array",
"items": {
"type": "string"
},
"description": "id or id@version"
},
"policies": {
"type": "object",
"properties": {
"maxMonthlyUsd": {
"type": "number",
"minimum": 0
},
"requireVerifiedOutcome": {
"type": "boolean"
}
}
},
"cursor": {
"type": "string",
"description": "From your last brief"
},
"stackFingerprint": {
"type": "string",
"description": "sha256:<64 hex>, hashed locally"
}
}
}🟢rescue(q, sku, provider, status, jtbd)
Diagnose a failing third-party API call before you retry or swap: paste the exact error (429, 402, 529 overloaded, insufficient_quota, model_not_found, deprecated SDK) and provider id. Returns the cause, an immediate fix and a fallback route, or no match plus checks to run. Not for picking a provider (decide).
输入模式
{
"type": "object",
"properties": {
"q": {
"type": "string",
"description": "Error text, status or SKU"
},
"sku": {
"type": "string",
"description": "Model or SKU id"
},
"provider": {
"type": "string",
"description": "Provider id that failed"
},
"status": {
"type": "string",
"description": "HTTP status, e.g. 429"
},
"jtbd": {
"type": "string",
"description": "Job id"
}
}
}🟢decide(task, jtbd, constraints, existing, maxMonthly, ...)
Choose a provider before writing integration code: LLM API, code sandbox, vector DB, web search, email, Postgres, auth, browser, observability. Send task in words or a jtbd id. Returns the dated editorial Depot Pick, a runner-up, the pick's price with source URL and date, and when not to take it. For quotes, comparisons and audits use depot.
输入模式
{
"type": "object",
"properties": {
"task": {
"type": "string",
"description": "What you need, in words"
},
"jtbd": {
"type": "string",
"description": "Or a job id, e.g. agent-sandbox; wins over task"
},
"constraints": {
"type": "string",
"description": "Comma list, e.g. eu-region; may switch the pick"
},
"existing": {
"type": "string",
"description": "Comma list of provider ids you already use"
},
"maxMonthly": {
"type": "number",
"minimum": 0,
"description": "USD budget for autoActOk"
},
"full": {
"type": "boolean"
}
}
}🟢depot(op, args)
Lookups beyond one pick; one op per call, args as an object: quote(jtbd) sourced SKU prices; estimate(jtbd, workload) monthly cost; compare(providers); free(jtbd) $0/OSS rows; check(providers) deprecated/retired flags; audit_repo(packages, models); stack(task), bundle(jtbds); search(q); kit(class); answer(task); stack_delta(providers, since); changes(since); failures; coverage; freshness. op=help gives each op's args.
输入模式
{
"type": "object",
"properties": {
"op": {
"type": "string",
"enum": [
"answer",
"estimate",
"search",
"stack",
"quote",
"bundle",
"kit",
"compare",
"free",
"failures",
"coverage",
"changes",
"check",
"freshness",
"audit_repo",
"stack_delta",
"help"
]
},
"args": {
"type": "object"
}
},
"required": [
"op"
]
}社区
证据