How to build a content calendar with Grok

Set up a content calendar for me in databasedmcp, in one go.

1. Create a database called content_calendar, described as "Content calendar: planned and published posts by channel and date, kept by Grok."
2. Create two tables:
   - campaigns: id, name (required), starts_on (date), ends_on (date), goal.
   - posts: id, title (required), channel (blog, newsletter, linkedin, x, instagram or youtube, required), status (idea, drafting, ready or published, defaults to idea), publish_on (the planned date, can be empty for ideas), campaign_id (references campaigns), notes, url (set once it's live), created_at (defaults to now).
3. Write the database a guide that says what each table is for, that a post is one channel so a blog post and the LinkedIn post about it are two rows, that ideas may have no date, that nothing here publishes anything and a post is marked published only when I say it's out, and the queries for "what's planned in the next 14 days" and "ideas without a date".

From then on: when we come up with a post, add it with its channel, a date if we have one and its campaign if it belongs to one. When I start a launch or theme, add it as a campaign. When I say a post is live, mark it published and save its link. Show me the tables when you're finished.

CREATE TABLE campaigns (
  id INTEGER PRIMARY KEY,
  name TEXT NOT NULL,
  starts_on TEXT,
  ends_on TEXT,
  goal TEXT
);

CREATE TABLE posts (
  id INTEGER PRIMARY KEY,
  title TEXT NOT NULL,
  channel TEXT NOT NULL CHECK (channel IN ('blog', 'newsletter', 'linkedin', 'x', 'instagram', 'youtube')),
  status TEXT NOT NULL DEFAULT 'idea' CHECK (status IN ('idea', 'drafting', 'ready', 'published')),
  publish_on TEXT,
  campaign_id INTEGER REFERENCES campaigns(id),
  notes TEXT,
  url TEXT,
  created_at TEXT NOT NULL DEFAULT CURRENT_TIMESTAMP
);

# content_calendar: posts and campaigns

- **campaigns**: a launch or theme that several posts belong to, with start and end dates.
- **posts**: one row per post on one channel. A blog post and the LinkedIn post about it are two rows. status: idea, drafting, ready, published. `publish_on` is the planned date; ideas may have none. `url` is set once it's live.

Next 14 days: `SELECT publish_on, channel, title, status FROM posts WHERE publish_on BETWEEN date('now') AND date('now', '+14 days') ORDER BY publish_on`

Ideas without a date: `SELECT id, title, channel FROM posts WHERE status = 'idea' AND publish_on IS NULL`

Nothing here publishes anything. The human posts; mark it published when they say so.

INSERT INTO campaigns (name, starts_on, ends_on, goal) VALUES (?, date('now'), date('now', '+21 days'), ?) RETURNING id
-- params: ["Spring collection","Newsletter sign-ups from the launch"]

INSERT INTO posts (title, channel, status, publish_on, campaign_id) VALUES (?, 'blog', 'drafting', date('now', 'weekday 2'), ?), (?, 'linkedin', 'idea', date('now', 'weekday 2', '+1 day'), ?)
-- params: ["How the spring collection was made",1,"Behind the spring collection, link to the blog",1]

INSERT INTO posts (title, channel) VALUES (?, ?)
-- params: ["Reader Q&A","newsletter"]

UPDATE posts SET status = 'published', url = ? WHERE id = ?
-- params: ["https://example.com/blog/spring-collection",1]

SELECT publish_on, channel, title, status FROM posts WHERE publish_on BETWEEN date('now') AND date('now', '+14 days') ORDER BY publish_on, channel

WITH RECURSIVE days(d) AS (SELECT date('now') UNION ALL SELECT date(d, '+1 day') FROM days WHERE d < date('now', '+13 days')) SELECT d AS free_day FROM days WHERE strftime('%w', d) NOT IN ('0', '6') AND d NOT IN (SELECT publish_on FROM posts WHERE publish_on IS NOT NULL) ORDER BY d

SELECT p.status, count(*) AS posts, group_concat(p.channel, ', ') AS channels FROM posts p JOIN campaigns c ON c.id = p.campaign_id WHERE c.name LIKE ? GROUP BY p.status
-- params: ["%spring%"]