Yu-Yang-Li/StarWhisper

近邻星系巡天（NGSS）的观测自动化代码，对应论文 [StarWhisper Telescope](https://doi.org/10.1038/s44172-025-00520-4)。程序负责制定观测计划、管理目标、连接 NINA，以及传输和处理观测数据。

- **计算模块与本地数据。** `PlanObservation3.py` 引用的 `src/module/observable_calculator.py` 未收录；`data/` 下的台站表、`50_bright_03.csv` 星表、按日期保存的计划与配置也未附带。代码使用的 `de440s.bsp`、`finals2000A.all` 及其路径需另行核对。 - **设备与通信。** 安装 [NINA](https://nighttime-imaging.com/) 

- **设备与通信。** 安装 [NINA](https://nighttime-imaging.com/) 及 `FMoraes.NINA.SitesPlugin.dll`，配置望远镜、FTP 和 MQTT。`src/module/UdpConnect.py` 中的通信初始化需要按自己的网络补齐，并核对 `topic.yaml`。

- **数据处理环境。** 按 [X-OPSTEP](https://github.com/Shinehope-Chen/X-OPSTEP/blob/master/README.md) 说明准备 astrometry.net、ASTAP、Scamp、SWarp、HOTPANTS，以及观测图像与模板。

`observe.yml` 的环境名为 `observe`，其中保留了 Linux 平台的包构建版本；其他平台需要调整依赖。服务默认使用端口 8000，启动后可查看[本地 API 文档](http://127.0.0.1:8000/docs)。`--reload` 用于开发调试；服务启动成功仍需分别验证设备连接和观测流程。

新表需遵循[读表工具的字段说明](../../skills/starwhisper-snclock/SKILL.md)。完整预测和数据获取流程在 [SitianClaw](https://github.com/Yu-Yang-Li/SitianClaw) 的 `snc-*` 技能中维护，本目录未附带其模型与完整输入。

[项目目录](../README.md) · [研究文档](../../docs/research.md) · [结果工具说明](../../skills/starwhisper-varlen/SKILL.md) · [模型权重](https://huggingface.co/castor0705/sparse-lc-early-classification)

GPU 训练还需按本机 CUDA 环境选择匹配的 [PyTorch](https://pytorch.org/get-started/locally/)。不同模型的训练入口见 [train_models/README.md](train_models/README.md)。

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

"dry_run_request": {     "method": "POST",     "url": "https://giiisp.com/first/paper/searchArxivByTitle",

"method": "POST",     "url": "https://giiisp.com/first/paper/searchArxivByTitle",     "headers": {

"url": "https://giiisp.com/first/paper/searchArxivByTitle",

"接口": "/first/paper/searchArxivByTitle",       "论文": "Example Paper Title for End-to-End Dry Run; First Author, Second Author; 2025; arXiv",       "链接": "https://arxiv.org/abs/2501.01234",

"论文": "Example Paper Title for End-to-End Dry Run; First Author, Second Author; 2025; arXiv",       "链接": "https://arxiv.org/abs/2501.01234",       "摘要依据": "Simulated title or abstract evidence matching the supplied query.",

"链接": "https://arxiv.org/abs/2501.01234",

"doi": null,       "arxiv_id": "2501.01234",       "url": "https://arxiv.org/abs/2501.01234",

"arxiv_id": "2501.01234",       "url": "https://arxiv.org/abs/2501.01234",       "pdf_url": "https://arxiv.org/pdf/2501.01234",

"url": "https://arxiv.org/abs/2501.01234",

"pdf_url": "https://arxiv.org/pdf/2501.01234",

"候选论文": "Example Paper Title; First Author, Second Author; 2025; arXiv",     "证据字段": "title/abstract/arxiv_id",     "链接": "https://arxiv.org/abs/2501.01234",

"证据字段": "title/abstract/arxiv_id",     "链接": "https://arxiv.org/abs/2501.01234",     "状态": "待核验",

"链接": "https://arxiv.org/abs/2501.01234",

"pageNum": 1,     "pageSize": 10,     "expected_url": "https://giiisp.com/first/oaPaper/searchArticlesByQuery1",

"pageSize": 10,     "expected_url": "https://giiisp.com/first/oaPaper/searchArticlesByQuery1",     "expected_body": {

"expected_url": "https://giiisp.com/first/oaPaper/searchArticlesByQuery1",

"pageNum": 1,     "pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxivByAbstract",

"pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxivByAbstract",     "expected_body": {

"expected_url": "https://giiisp.com/first/paper/searchArxivByAbstract",

"pageNum": 1,     "pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxivByArxivNo1",

"pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxivByArxivNo1",     "expected_body": {

"expected_url": "https://giiisp.com/first/paper/searchArxivByArxivNo1",

"pageNum": 1,     "pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxiv",

"pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxiv",     "expected_body": {

"expected_url": "https://giiisp.com/first/paper/searchArxiv",

"pageNum": 1,     "pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxivByTitle",

"pageSize": 10,     "expected_url": "https://giiisp.com/first/paper/searchArxivByTitle",     "expected_body": {

"expected_url": "https://giiisp.com/first/paper/searchArxivByTitle",

基址为 `https://giiisp.com`。所有接口使用 `POST` 和 JSON body，不把业务参数放在 URL query。

1. 明确用户要查的主题、时间范围、领域和语言。 2. 把用户原话整理成 1-3 组检索词，不要替用户编论文名。 3. 如果接口需要登录态或鉴权，但当前没有凭据，只构造请求体和 curl 示例，不真实调用；提醒用户到 `https://giiisp.com/#/mcp/authenticate` 申请或刷新 Giiisp MCP 认证，并设置 `GIIISP_AUTH_TOKEN`。

2. 把用户原话整理成 1-3 组检索词，不要替用户编论文名。 3. 如果接口需要登录态或鉴权，但当前没有凭据，只构造请求体和 curl 示例，不真实调用；提醒用户到 `https://giiisp.com/#/mcp/authenticate` 申请或刷新 Giiisp MCP 认证，并设置 `GIIISP_AUTH_TOKEN`。 4. 如果返回 HTML 登录页、非 JSON、401 或 403，先报告鉴权问题和认证页，再给开放源回退方案。

[Credential-related source omitted]

[Credential-related source omitted]

|---|---| | `search_started` | 搜索开始，包含主检索模式、扩展模式和计划请求数 | | `auth_status` | 报告是否检测到 `GIIISP_AUTH_TOKEN`；无 token 时给出 `https://giiisp.com/#/mcp/authenticate` 和用户动作 |

| `search_started` | 搜索开始，包含主检索模式、扩展模式和计划请求数 | | `auth_status` | 报告是否检测到 `GIIISP_AUTH_TOKEN`；无 token 时给出 `https://giiisp.com/#/mcp/authenticate` 和用户动作 | | `request_prepared` | 某个接口/页码的请求体已生成，可展示“正在查 title / abstract / OA” |

[Credential-related source omitted]

[Credential-related source omitted]

"doi": null,       "arxiv_id": "2501.01234",       "url": "https://arxiv.org/abs/2501.01234",

"arxiv_id": "2501.01234",       "url": "https://arxiv.org/abs/2501.01234",       "pdf_url": "https://arxiv.org/pdf/2501.01234",

"url": "https://arxiv.org/abs/2501.01234",

"pdf_url": "https://arxiv.org/pdf/2501.01234",

"doi": null,       "arxiv_id": "2501.01234",       "url": "https://arxiv.org/abs/2501.01234",

"arxiv_id": "2501.01234",       "url": "https://arxiv.org/abs/2501.01234",       "pdf_url": "https://arxiv.org/pdf/2501.01234",

"url": "https://arxiv.org/abs/2501.01234",

"pdf_url": "https://arxiv.org/pdf/2501.01234",

Executable file

ADS_SEARCH = "https://api.adsabs.harvard.edu/v1/search/query"

ARXIV_SEARCH = "https://export.arxiv.org/api/query"

USER_AGENT = "StarWhisper-skills/ads-first (https://github.com/Yu-Yang-Li/StarWhisper)"

ARXIV_NS = {     "atom": "http://www.w3.org/2005/Atom",     "arxiv": "http://arxiv.org/schemas/atom",

"atom": "http://www.w3.org/2005/Atom",

"arxiv": "http://arxiv.org/schemas/atom",

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

def _get(url: str, headers: dict[str, str]) -> tuple[int, str, bytes]:     req = urllib.request.Request(url, headers=headers, method="GET")

req = urllib.request.Request(url, headers=headers, method="GET")

try:         with urllib.request.urlopen(req, timeout=30) as resp:             return resp.status, resp.headers.get("Content-Type", ""), resp.read()

with urllib.request.urlopen(req, timeout=30) as resp:

"doi": (item.get("doi") or [None])[0],                 "arxiv_id": arxiv_id,                 "url": f"https://ui.adsabs.harvard.edu/abs/{item.get('bibcode')}/abstract" if item.get("bibcode") else None,

"arxiv_id": arxiv_id,                 "url": f"https://ui.adsabs.harvard.edu/abs/{item.get('bibcode')}/abstract" if item.get("bibcode") else None,                 "verification_status": "ADS hit",

"url": f"https://ui.adsabs.harvard.edu/abs/{item.get('bibcode')}/abstract" if item.get("bibcode") else None,

"doi": doi,                 "arxiv_id": arxiv_id,                 "url": f"https://arxiv.org/abs/{arxiv_id}" if arxiv_id else None,

"arxiv_id": arxiv_id,                 "url": f"https://arxiv.org/abs/{arxiv_id}" if arxiv_id else None,                 "verification_status": "preprint",

"url": f"https://arxiv.org/abs/{arxiv_id}" if arxiv_id else None,

Executable file

BASE_URL = "https://giiisp.com"

"doi": None,                     "arxiv_id": "2501.01234",                     "url": "https://arxiv.org/abs/2501.01234",

"arxiv_id": "2501.01234",                     "url": "https://arxiv.org/abs/2501.01234",                     "pdf_url": "https://arxiv.org/pdf/2501.01234",

"url": "https://arxiv.org/abs/2501.01234",

"pdf_url": "https://arxiv.org/pdf/2501.01234",

"candidate_paper": "Example Paper Title for Dry Run; First Author, Second Author; 2025; arXiv",                 "evidence_field": "title/abstract",                 "link": "https://arxiv.org/abs/2501.01234",

"evidence_field": "title/abstract",                 "link": "https://arxiv.org/abs/2501.01234",                 "status": "待核验",

"link": "https://arxiv.org/abs/2501.01234",

"doi": None,                     "arxiv_id": "2501.01234",                     "url": "https://arxiv.org/abs/2501.01234",

"arxiv_id": "2501.01234",                     "url": "https://arxiv.org/abs/2501.01234",                     "pdf_url": "https://arxiv.org/pdf/2501.01234",

"url": "https://arxiv.org/abs/2501.01234",

"pdf_url": "https://arxiv.org/pdf/2501.01234",

"candidate_paper": "Example Open Source Paper Title for Fallback; First Author, Second Author; 2025; arXiv",                 "evidence_field": "open source title/abstract/identifier metadata",                 "link": "https://arxiv.org/abs/

"evidence_field": "open source title/abstract/identifier metadata",                 "link": "https://arxiv.org/abs/2501.01234",                 "status": "非 Giiisp 结果",

"link": "https://arxiv.org/abs/2501.01234",

"接口": path,                     "论文": "Example Paper Title for End-to-End Dry Run; First Author, Second Author; 2025; arXiv",                     "链接": "https://arxiv.org/abs/2501.01234",

"论文": "Example Paper Title for End-to-End Dry Run; First Author, Second Author; 2025; arXiv",                     "链接": "https://arxiv.org/abs/2501.01234",                     "摘要依据": "Simulated title or abstract evidence matching the suppl

"链接": "https://arxiv.org/abs/2501.01234",

#!/usr/bin/env python3

BASE_URL = "https://giiisp.com"

GIIISP_AUTH_URL = "https://giiisp.com/#/mcp/authenticate"

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

def post_json(request_payload: dict[str, Any], timeout: float) -> dict[str, Any]:     data = json.dumps(request_payload["body"], ensure_ascii=False).encode("utf-8")     request = urllib.request.Request(

data = json.dumps(request_payload["body"], ensure_ascii=False).encode("utf-8")     request = urllib.request.Request(         request_payload["url"],

request = urllib.request.Request(

)     try:         with urllib.request.urlopen(request, timeout=timeout) as response:

try:         with urllib.request.urlopen(request, timeout=timeout) as response:             raw = response.read().decode("utf-8", errors="replace")

with urllib.request.urlopen(request, timeout=timeout) as response:

Executable file

assert "no papers were invented" in payload["safety"]     ads = payload["planned_requests"]["ads"]     assert ads["url"] == "https://api.adsabs.harvard.edu/v1/search/query"

ads = payload["planned_requests"]["ads"]     assert ads["url"] == "https://api.adsabs.harvard.edu/v1/search/query"     assert "StarWhisper Telescope NGSS" in ads["params"]["q"]

assert ads["url"] == "https://api.adsabs.harvard.edu/v1/search/query"

Executable file

payload = run_dry(mode)         assert payload["method"] == "POST"         assert payload["url"] == "https://giiisp.com" + path

assert payload["method"] == "POST"         assert payload["url"] == "https://giiisp.com" + path         assert payload["body"] == body

assert payload["url"] == "https://giiisp.com" + path

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

assert events[1]["provider"] == "giiisp"     assert events[1]["ok"] is False     assert events[1]["auth_url"] == "https://giiisp.com/#/mcp/authenticate"

assert events[1]["ok"] is False     assert events[1]["auth_url"] == "https://giiisp.com/#/mcp/authenticate"     assert events[1]["user_action"] == "申请或刷新 Giiisp MCP 认证后设置 GIIISP_AUTH_TOKEN"

assert events[1]["auth_url"] == "https://giiisp.com/#/mcp/authenticate"

[Credential-related source omitted]

[Credential-related source omitted]

assert events[2]["request"]["url"] == "https://giiisp.com/first/paper/searchArxivByTitle"

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

| 项 | 内容 | |---|---| | 根页 | `http://images.sitianai.com/` |

|---|---| | 根页 | `http://images.sitianai.com/` | | 生成 | `POST http://images.sitianai.com/api/generate-async` |

| 根页 | `http://images.sitianai.com/` |

| 生成 | `POST http://images.sitianai.com/api/generate-async` |

[Credential-related source omitted]

- 只记录“已使用访问码 token”。 - 命令行测试只从环境变量 `GIIISP_AUTH_TOKEN` 读取 token。 - 没有 token 时提醒用户到 `https://giiisp.com/#/mcp/authenticate` 申请或刷新 Giiisp 认证，再设置 `GIIISP_AUTH_TOKEN`。

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

Executable file

Executable file

[Credential-related source omitted]

Executable file

ENDPOINT = "http://images.sitianai.com/api/generate-async"

"headers": {             "Content-Type": "application/json",             "Referer": "http://images.sitianai.com/",

[Credential-related source omitted]

"Referer": "http://images.sitianai.com/",

[Credential-related source omitted]

Executable file

import base64 import hashlib import http.client

import hashlib import http.client import json

import http.client

ROOT = "http://images.sitianai.com/"

[Credential-related source omitted]

except TimeoutError as exc:         return {"status_code": None, "json": {"error": str(exc), "type": "TimeoutError"}, "text": str(exc)}     except (http.client.HTTPException, ConnectionError, OSError) as exc:

except (http.client.HTTPException, ConnectionError, OSError) as exc:

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

data = base64.b64decode(encoded)

data = base64.b64decode(candidate["value"])

print("BLOCKED: " + reason)     if "GIIISP_AUTH_TOKEN" in reason or "ACCESS_TOKEN_REQUIRED" in reason:         print("Apply for or refresh Giiisp authentication at https://giiisp.com/#/mcp/authenticate")

[Credential-related source omitted]

[Credential-related source omitted]

print("Apply for or refresh Giiisp authentication at https://giiisp.com/#/mcp/authenticate")

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

Executable file

import argparse import base64 import http.client

Executable file

import base64 import http.client import json

import http.client

ENDPOINT = "https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions"

[Credential-related source omitted]

[Credential-related source omitted]

data = {"raw_text": text}             return {"status_code": exc.code, "json": data, "text": text}         except (URLError, TimeoutError, http.client.HTTPException, ConnectionError, OSError) as exc:

except (URLError, TimeoutError, http.client.HTTPException, ConnectionError, OSError) as exc:

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

print(             "Apply for a DashScope/Bailian API key at https://help.aliyun.com/zh/model-studio/get-api-key",             file=sys.stderr,

"Apply for a DashScope/Bailian API key at https://help.aliyun.com/zh/model-studio/get-api-key",

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

Executable file

[Credential-related source omitted]

Executable file

+ b"\x08\x02\x00\x00\x00"

+ b"\x00\x00\x00\x00"

Executable file

payload = run_dry("--caption", "科研流程图", "--intent", "展示从问题到结果的流程")     assert payload["method"] == "POST"     assert payload["url"] == "http://images.sitianai.com/api/generate-async"

assert payload["method"] == "POST"     assert payload["url"] == "http://images.sitianai.com/api/generate-async"     assert payload["body"]["responseModalities"] == ["IMAGE", "TEXT"]

assert payload["url"] == "http://images.sitianai.com/api/generate-async"

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

+ b"\x08\x02\x00\x00\x00"

+ b"\x00\x00\x00\x00"

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

This is the citation-checking subset of TaShan-PaperChecker distributed with the [upstream research skills](https://github.com/TashanGKD/tashan-research-skills).

the [upstream research skills](https://github.com/TashanGKD/tashan-research-skills).

[Credential-related source omitted]

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

[Credential-related source omitted]

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

self.base_url = "https://mineru.net/api/v4"

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

response = requests.post(url, headers=self.headers, json=data)

response = requests.get(url, headers=self.headers)

response = requests.post(url, headers=self.headers, json=data)

with open(file_path, 'rb') as f:             response = requests.put(upload_url, data=f)

response = requests.put(upload_url, data=f)

response = requests.get(url, headers=self.headers)

print(f"Downloading result from {zip_url}...")             response = requests.get(zip_url, stream=True)

response = requests.get(zip_url, stream=True)

[Credential-related source omitted]

#!/usr/bin/env python

[Credential-related source omitted]

"pymupdf4llm", ) MINERU_AUTH_URL = "https://mineru.net"

MINERU_AUTH_URL = "https://mineru.net"

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python

#!/usr/bin/env python

SKILL_ROOT = Path(__file__).resolve().parents[1] RULES_ROOT = SKILL_ROOT / "assets" / "paperchecker-rules" AUTH_URL = "https://mineru.net"

RULES_ROOT = SKILL_ROOT / "assets" / "paperchecker-rules" AUTH_URL = "https://mineru.net" NUMBERED_CITATION_RE = re.compile(r"\[(\d+)(?:\s*-\s*(\d+))?\]")

AUTH_URL = "https://mineru.net"

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python

[Credential-related source omitted]

user_action = env_report.get("pdf_extraction", {}).get("user_action") or ""             if not env_report.get("pdf_extraction", {}).get("configured"):                 fail_if("https://mineru.net" not in user_action, "MinerU action must ment

if not env_report.get("pdf_extraction", {}).get("configured"):                 fail_if("https://mineru.net" not in user_action, "MinerU action must mention https://mineru.net", failures)         except json.JSONDecodeError:

fail_if("https://mineru.net" not in user_action, "MinerU action must mention https://mineru.net", failures)

if smoke_input.suffix.lower() == ".pdf":             fail_if(                 "未配置 MinerU API key，请到 https://mineru.net 申请" not in smoke.stdout,

fail_if(                 "未配置 MinerU API key，请到 https://mineru.net 申请" not in smoke.stdout,                 "PDF dry-run stream missing MinerU action",

"未配置 MinerU API key，请到 https://mineru.net 申请" not in smoke.stdout,

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

Executable file

[Credential-related source omitted]

Executable file

#!/usr/bin/env python

#!/usr/bin/env python

Executable file

emit(f"[设备] {DEVICE}", log_lines)     tr_ld, va_ld, auto_nc = make_loaders(args.data_dir, args.batch_size, CONFIG["num_workers"], log_lines)     nc = args.num_classes or auto_nc

tr_ld, va_ld, auto_nc = make_loaders(args.data_dir, args.batch_size, CONFIG["num_workers"], log_lines)     nc = args.num_classes or auto_nc     model = build_model(nc).to(DEVICE)

nc = args.num_classes or auto_nc

Executable file

Executable file

Executable file

POST http://123.56.218.60:18000/api/research/ask

#!/usr/bin/env python3

#!/usr/bin/env python3

DEFAULT_ENDPOINT = "http://123.56.218.60:18000/api/research/ask"

AUTH_URL = "https://giiisp.com/#/mcp/authenticate"

if not health_url:         return False, interface_unavailable_event("missing_health_url")     request = urllib.request.Request(health_url, method="GET")

return False, interface_unavailable_event("missing_health_url")     request = urllib.request.Request(health_url, method="GET")     try:

request = urllib.request.Request(health_url, method="GET")

try:         with urllib.request.urlopen(request, timeout=timeout) as response:             status = getattr(response, "status", response.getcode())

with urllib.request.urlopen(request, timeout=timeout) as response:

def lines_from_http(endpoint: str, body: dict[str, Any], timeout: float) -> Iterator[str]:     request = urllib.request.Request(

request = urllib.request.Request(

)     try:         with urllib.request.urlopen(request, timeout=timeout) as response:

try:         with urllib.request.urlopen(request, timeout=timeout) as response:             for raw in response:

with urllib.request.urlopen(request, timeout=timeout) as response:

Executable file

"",                 "event: references",                 'data: {"references":[{"title":"Paper A","url":"https://example.test/a","sourceApi":"searchArxiv"}]}',

"event: references",                 'data: {"references":[{"title":"Paper A","url":"https://example.test/a","sourceApi":"searchArxiv"}]}',                 "",

'data: {"references":[{"title":"Paper A","url":"https://example.test/a","sourceApi":"searchArxiv"}]}',

]     assert events[-1]["answer_status"] == "interface_unavailable"     assert "https://giiisp.com/#/mcp/authenticate" in events[-1]["user_action"]

assert "https://giiisp.com/#/mcp/authenticate" in events[-1]["user_action"]

Deep mode should cite this method paper at the start when relevant: https://link.springer.com/article/10.1140/epjds/s13688-026-00672-z. Follow its practical workflow shape: retrieve literature, combine abstracts with compressed full-text re

#!/usr/bin/env python

#!/usr/bin/env python

ATOM = {"atom": "http://www.w3.org/2005/Atom"}

}     )     url = "https://export.arxiv.org/api/query?" + params

)     url = "https://export.arxiv.org/api/query?" + params     with urllib.request.urlopen(url, timeout=30) as response:

url = "https://export.arxiv.org/api/query?" + params     with urllib.request.urlopen(url, timeout=30) as response:         root = ET.fromstring(response.read())

with urllib.request.urlopen(url, timeout=30) as response:

Executable file

[Credential-related source omitted]

Executable file

计算使用 `scripts/plan_night.py`，只依赖 Python 标准库，不建立 HTTP、MQTT、FTP 或 NINA 连接。配置来自 [StarWhisper Telescope](https://doi.org/10.1038/s44172-025-00520-4) 对应的 [`projects/ngss/`](../../projects/ngss/README.md)，与 Explore 合成实验和虚拟司天分开使用。

计算使用 `scripts/plan_night.py`，只依赖 Python 标准库，不建立 HTTP、MQTT、FTP 或 NINA 连接。配置来自 [StarWhisper Telescope](https://doi.org/10.1038/s44172-025-00520-4) 对应的 [`projects/ngss/`](../../projects/ngss/README.md)，与 Explore 合成实验和虚拟司天分开使用。

计算使用 `scripts/plan_night.py`，只依赖 Python 标准库，不建立 HTTP、MQTT、FTP 或 NINA 连接。配置来自 [StarWhisper Telescope](https://doi.org/10.1038/s44172-025-00520-4) 对应的 [`projects/ngss/`](../../projects/ngss/README.md)，与 Explore 合成实验和虚拟司天分开使用。

Executable file

[Credential-related source omitted]

Executable file

抓取、强制测光、可见性和三日筛选工作流见 [SitianClaw](https://github.com/Yu-Yang-Li/SitianClaw) 的 `snc-*` 技能。观测配置与曝光预算检查见 [`starwhisper-night-plan`](../starwhisper-night-plan/SKILL.md)。

抓取、强制测光、可见性和三日筛选工作流见 [SitianClaw](https://github.com/Yu-Yang-Li/SitianClaw) 的 `snc-*` 技能。观测配置与曝光预算检查见 [`starwhisper-night-plan`](../starwhisper-night-plan/SKILL.md)。

抓取、强制测光、可见性和三日筛选工作流见 [SitianClaw](https://github.com/Yu-Yang-Li/SitianClaw) 的 `snc-*` 技能。观测配置与曝光预算检查见 [`starwhisper-night-plan`](../starwhisper-night-plan/SKILL.md)。

Executable file

[Credential-related source omitted]

Executable file

- 测试集指标不是爆发时刻，也不是发现。 - 不要拿 StarWhisper LC 的 Kepler/K2 约 90% 顶替本基准。 - 权重见 [Hugging Face](https://huggingface.co/castor0705/sparse-lc-early-classification)；本技能不下载权重或执行模型推理。

- 权重见 [Hugging Face](https://huggingface.co/castor0705/sparse-lc-early-classification)；本技能不下载权重或执行模型推理。

Executable file

Executable file

#!/usr/bin/env python3

#!/usr/bin/env python3

The MinerU script reads the token from `MINERU_API_TOKEN`, hidden TTY input, or stdin. If the user has no token, point them to `https://mineru.net/apiManage/token` to register and create one. Never write the token into a command, report, or

#!/usr/bin/env python3

#!/usr/bin/env python3

NS = {     "w": "http://schemas.openxmlformats.org/wordprocessingml/2006/main",

"w": "http://schemas.openxmlformats.org/wordprocessingml/2006/main",

"m": "http://schemas.openxmlformats.org/officeDocument/2006/math",

"a": "http://schemas.openxmlformats.org/drawingml/2006/main",

"r": "http://schemas.openxmlformats.org/officeDocument/2006/relationships",

#!/usr/bin/env python3

#!/usr/bin/env python3

API_BASE = "https://mineru.net/api/v4"

[Credential-related source omitted]

[Credential-related source omitted]

return token     print("No MINERU_API_TOKEN found. Apply for a token at https://mineru.net/apiManage/token", file=sys.stderr)     if sys.stdin.isatty():

print("No MINERU_API_TOKEN found. Apply for a token at https://mineru.net/apiManage/token", file=sys.stderr)

"model_version": model_version,     }     resp = requests.post(f"{API_BASE}/file-urls/batch", headers=make_headers(token), data=json.dumps(payload), timeout=60)

}     resp = requests.post(f"{API_BASE}/file-urls/batch", headers=make_headers(token), data=json.dumps(payload), timeout=60)     print("create_status", resp.status_code)

resp = requests.post(f"{API_BASE}/file-urls/batch", headers=make_headers(token), data=json.dumps(payload), timeout=60)

def upload_file(url: str, file_path: Path) -> None:     with file_path.open("rb") as fh:         resp = requests.put(url, data=fh, timeout=600)

with file_path.open("rb") as fh:         resp = requests.put(url, data=fh, timeout=600)     print("upload_status", resp.status_code)

resp = requests.put(url, data=fh, timeout=600)

last_state = None     while time.time() < deadline:         resp = requests.get(f"{API_BASE}/extract-results/batch/{batch_id}", headers=make_headers(token), timeout=60)

while time.time() < deadline:         resp = requests.get(f"{API_BASE}/extract-results/batch/{batch_id}", headers=make_headers(token), timeout=60)         resp.raise_for_status()

resp = requests.get(f"{API_BASE}/extract-results/batch/{batch_id}", headers=make_headers(token), timeout=60)

out_dir.mkdir(parents=True, exist_ok=True)     zip_path = out_dir / safe_zip_name(zip_url, f"{fallback_name}.zip")     with requests.get(zip_url, stream=True, timeout=600) as resp:

zip_path = out_dir / safe_zip_name(zip_url, f"{fallback_name}.zip")     with requests.get(zip_url, stream=True, timeout=600) as resp:         resp.raise_for_status()

with requests.get(zip_url, stream=True, timeout=600) as resp:

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

NS = {"w": "http://schemas.openxmlformats.org/wordprocessingml/2006/main"}

#!/usr/bin/env python3

#!/usr/bin/env python3

[Credential-related source omitted]

[Credential-related source omitted]

- Codex `image_gen` if available in chat. - Giiisp/SiTian Imagine when `GIIISP_AUTH_TOKEN` is set. If the user has no token, point them to `https://giiisp.com/#/mcp/authenticate` to apply for or refresh Giiisp authentication.

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

- Generate: `POST http://images.sitianai.com/api/generate-async`

- Poll: `GET http://images.sitianai.com/api/generate-jobs/{job_id}`

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

Executable file

#!/usr/bin/env python3

import argparse import http.client

import http.client

ENDPOINT = "https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions"

[Credential-related source omitted]

[Credential-related source omitted]

"text": str(exc),         }     except (ConnectionError, http.client.HTTPException, TimeoutError, OSError) as exc:

}     except (ConnectionError, http.client.HTTPException, TimeoutError, OSError) as exc:         return {"status_code": None, "json": {"error": str(exc), "type": type(exc).__name__}, "text": str(exc)}

except (ConnectionError, http.client.HTTPException, TimeoutError, OSError) as exc:

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

blockers.append(             "no auditable image-editing backend is available in the local runtime; "             "apply for GIIISP_AUTH_TOKEN at https://giiisp.com/#/mcp/authenticate "

"no auditable image-editing backend is available in the local runtime; "             "apply for GIIISP_AUTH_TOKEN at https://giiisp.com/#/mcp/authenticate "             "or use the host platform's built-in image generation"

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python3

#!/usr/bin/env python3

Executable file

[Credential-related source omitted]

[Credential-related source omitted]

args = parser.parse_args()     if not os.environ.get("GIIISP_AUTH_TOKEN"):         raise SystemExit("GIIISP_AUTH_TOKEN is not set; authenticate at https://giiisp.com/#/mcp/authenticate if the token is missing or expired")

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python3

Executable file

#!/usr/bin/env python3

import argparse import http.client

import http.client

ENDPOINT = "https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions"

[Credential-related source omitted]

[Credential-related source omitted]

"text": str(exc),         }     except (ConnectionError, http.client.HTTPException, TimeoutError, OSError) as exc:

}     except (ConnectionError, http.client.HTTPException, TimeoutError, OSError) as exc:         return {"status_code": None, "json": {"error": str(exc), "type": type(exc).__name__}, "text": str(exc)}

except (ConnectionError, http.client.HTTPException, TimeoutError, OSError) as exc:

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3