{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_jgesyqhs7puk","handle":"annual-series-forecast-mcp","url":"https://wellknown.network/agents/annual-series-forecast-mcp","links":{"self":"https://wellknown.network/agents/annual-series-forecast-mcp/record.json","html":"https://wellknown.network/agents/annual-series-forecast-mcp","markdown":"https://wellknown.network/agents/annual-series-forecast-mcp/record.md","api":"https://wellknown.network/api/v1/agents/annual-series-forecast-mcp","status":"https://wellknown.network/api/v1/agents/annual-series-forecast-mcp/status","claim":"https://wellknown.network/agents/annual-series-forecast-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/annual-series-forecast-mcp/claim.json","badge":"https://wellknown.network/agents/annual-series-forecast-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/annual-series-forecast-mcp/history","tools":"https://wellknown.network/api/v1/agents/annual-series-forecast-mcp/tools"},"ard":{"identifier":"urn:air::server:annual-series-forecast-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"annual-series-forecast-mcp","summary":"年度时间序列预测 MCP - 基于 ARIMA / 指数平滑的同比增速与水平值联合预测","description":"# Annual Series Forecast MCP \n \n年度时间序列预测 MCP (Model Context Protocol)。基于 ARIMA 与 Holt 指数平滑，对按年排列的序列做同比增速与水平值联合预测。 \n \n适用于含 `year`、`growth_rate`（同比增速%）及可选 `value`（水平绝对值）的年度数据，例如宏观经济指标、行业统计等。测试用例中的 GDP 等指标只是典型场景之一。 \n \n与 `time-series-forecast-mcp` 的区别：本服务面向「年度 + 增速/水平值」结构化输入，内置模型族选择与近年加权均值组合；后者面向原始数值序列，提供 AR/MA/ARIMA/GARCH/指数平滑等多种单模型。 \n \n## 安装与运行 \n \n### 方式一：uvx（推荐，无需手动安装） \n \n需先安装 [uv](https://docs.astral.sh/uv/)： \n \n```bash \n# 本地试跑（stdio MCP 服务） \nuvx annual-series-forecast-mcp \n``` \n \n发布到 PyPI 后，其他机器可直接用包名运行；未发布前可指定源码路径： \n \n```bash \nuvx --from /path/to/annual-series-forecast-mcp annual-series-forecast-mcp \n``` \n \n### 方式二：pip 安装 \n \n```bash \npip install annual-series-forecast-mcp \nannual-series-forecast-mcp \n``` \n \n### 方式三：源码开发 \n \n```bash \ncd annual-series-forecast-mcp \npip install -e \".[test]\" \npython -m annual_series_forecast_mcp \n``` \n \n### Cursor / Claude Desktop 配置（uvx） \n \n配置文件路径示例： \n- Windows: `%USERPROFILE%\\.cursor\\mcp.json` \n- macOS/Linux: `~/.cursor/mcp.json` \n \n```json \n{ \n  \"mcpServers\": { \n    \"annual-series-forecast\": { \n      \"command\": \"uvx\", \n      \"args\": [\"annual-series-forecast-mcp\"] \n    } \n  } \n} \n``` \n \n若使用本地源码目录（未发布到 PyPI）： \n \n```json \n{ \n  \"mcpServers\": { \n    \"annual-series-forecast\": { \n      \"command\": \"uvx\", \n      \"args\": [ \n        \"--from\", \n        \"C:/local/code/ai-base-3/annual-series-forecast-mcp\", \n        \"annual-series-forecast-mcp\" \n      ] \n    } \n  } \n} \n``` \n \n### 使用 python -m 的配置示例 \n \n```json \n{ \n  \"mcpServers\": { \n    \"annual-series-forecast\": { \n      \"command\": \"python\", \n      \"args\": [\"-m\", \"annual_series_forecast_mcp\"] \n    } \n  } \n} \n``` \n \n### 工具列表 \n \n#### forecast_next_year \n \n基于历史年度时间序列预测下一期。 \n \n**参数**: \n \n- `history`: 历史记录列表（按年份升序），每项含 `year`；有水平值时含 `value`，否则含 `growth_rate` \n- `confidence_level`: 置信水平，默认 0.95 \n- `use_trend_blend`: 是否在结构模型与近年加权均值间做组合，默认 true \n- `recent_window`: 近年加权均值窗口（年），默认 7 \n- `holdout_years`: 滚动交叉验证留出年数，默认 5 \n- `forecast_level`: 是否预测水平值；`null` 时根据最近一期是否有 `value` 自动判断 \n- `p`, `d`, `q`: 手动指定 ARIMA …","publisher":null,"homepage":"https://github.com/your-org/annual-series-forecast-mcp","repository":"https://github.com/your-org/annual-series-forecast-mcp","version":"0.1.0","license":"MIT","protocols":["mcp"],"tags":["time-series","forecasting","mcp","arima","exponential-smoothing","annual"],"pricing":null,"endpoints":[{"url":"pypi:annual-series-forecast-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"data.weather","name":"Weather","confidence":1,"provenance":"derived"}],"categories":["data"],"language":"es"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"annual-series-forecast-mcp","registry":"pypi","observedAt":"2026-11-08T23:27:18.132Z","publishedAt":"2026-06-11T09:24:06.747864Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"annual-series-forecast-mcp","url":"https://pypi.org/project/annual-series-forecast-mcp/","firstSeenAt":"2026-09-09T09:22:54.121Z","fetchedAt":"2026-11-08T23:26:00.568Z","normalizedAt":"2026-11-08T23:26:00.568Z"}]},"firstSeenAt":"2026-09-09T09:22:54.121Z","updatedAt":"2026-11-08T23:27:18.132Z"}