daily-news

Daily News

Three-stage workflow:Fetch metadata → Generate summaries → Output daily digest

Working directory

On first run, ask for the working directory path (e.g. ~/daily-news), then remember it.

<workspace>/
├── profile.yaml          # 用户画像（关于我、关注什么）
├── settings.yaml         # 日报设置（语言、格式偏好）
├── methods/              # 信源获取方法
├── data/news.db          # SQLite 数据库
└── output/YYYY-MM-DD.md  # 日报输出

Initialization:

mkdir -p <workspace>/methods <workspace>/data <workspace>/output
cp references/examples/settings.example.yaml <workspace>/settings.yaml
cp references/examples/profile.example.yaml <workspace>/profile.yaml
python3 scripts/db.py init --db <workspace>/data/news.db

After initialization completes, write the working directory to ~/.claude/CLAUDE.md：

- daily-news skill 的项目目录在：<workspace>

Stage 1: Fetch metadata

Iterate over the <workspace>/methods/ directory, and run the fetch for each method file.

Incremental fetching: each source independently tracks last_fetched_date, which exists in two places: the method YAML and the DB. Default is incremental mode; on the first fetch, ask the user for the date range.

python3 scripts/db.py source-status --db <db> --source <source_id>

Deduplication: three layers of protection — filter by since during fetching → check-existing pre-check before inserting into the DB → DB UNIQUE constraint as a fallback.

Method execution：

# extends: rss
python3 references/methods/rss.py --url "<source_url>"

# extends: webfetch-smart 或 browser-smart
# 遵循 web-access skill 进行联网操作

# 无 extends：*.py 直接执行，*.md 按内容操作

Insert into DB:

python3 scripts/db.py add-items-incremental \
  --db <db> --source <source_id> --items '<json>' --since "YYYY-MM-DD"

Stage 2: Generate summaries

python3 scripts/db.py list-pending --db <workspace>/data/news.db

For each item, fetch the body according to the method's detail_method field (follow the web-access skill), and generate a summary according to references/prompts/summary.md :

python3 scripts/db.py update-summary --db <db> --id <item_id> --data '<摘要JSON>'

Stage 3: Generate daily digest

python3 scripts/db.py list-today --db <workspace>/data/news.db

Read profile.yaml, generate the daily digest according to references/prompts/report.md , and output to output/YYYY-MM-DD.md。

User profile

profile.yaml format:

about: |
  （关于我：身份、工作）

focus: |
  （当前关注：最近在意的话题）

low_priority: |
  （不太关心：降低优先级的内容）

Add source

Try in order of priority:

Check RSS — try /feed、/rss、/atom.xml, or the page <link rel="alternate">, if present use extends: rss
WebFetch or browser — follow the web-access skill, if reachable use extends: webfetch-smart, otherwise extends: browser-smart

For creating a method file, see references/schemas/method.md

Reference materials

Material	Path	Load timing
Method specification	`references/schemas/method.md`	When adding a source
Summary prompt	`references/prompts/summary.md`	Stage 2
Daily digest prompt	`references/prompts/report.md`	Stage 3
RSS method	`references/methods/rss.py`	Stage 1 (`extends: rss`）
Method metadata example	`references/examples/method-with-metadata.example.yaml`	When adding a source / configuring incremental tracking
Website deployment	`references/website-deployment.md`	When the user requests deployment

Dependencies

pip install pyyaml feedparser requests beautifulsoup4