# Bl Week

> Plan the week on LinkedIn - what to post, when to post it, and who to engage with. Use when the user says "plan my week", "what should I post", "content calendar", "I have nothing to post about", or wants a posting schedule and an engagement list.

- Skill: `ehteshambuildagents/bl-week` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ehteshambuildagents/bl-week`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ehteshambuildagents/bl-week/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- Author: ehteshambuildagents (https://skillmd.com/u/ehteshambuildagents)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/ehteshambuildagents/bl-week

---


# bl-week

The control room. Every other skill in this pack executes; this one decides what
gets executed. Run it once a week, on the same day, or it stops working.

## Input

Read `~/.claude/bylinekit/voice.md`, `ledger.md`, and `log.md` if it exists. The
plan must not repeat a theme from the last fortnight, and `log.md` is how you
know what those were.

Then ask for one thing, and only one: **what actually happened this week.** A
client call, a number that moved, a mistake, a thing they built, an argument
they had, a question they got asked twice. This is where posts come from. A
week with nothing in it produces a week of generic posts, and the honest answer
then is to post less, not to invent more.

## The week

Four posts beats seven. Seven mediocre posts train the feed to skip the user,
and they cost more than they return.

| day | slot | shape |
| --- | --- | --- |
| Monday | the opinion | the position they want to be known for |
| Tuesday | - | engagement only |
| Wednesday | the proof | a number, a case, a result from the ledger |
| Thursday | the story | something that happened, told properly |
| Friday | - | engagement only |
| optional | the artifact | a template, a checklist, a thing to take |

Post when the user's audience is at a desk, which is usually mid-morning in
their timezone. Do not chase a mythical optimal minute. Consistency beats
timing by a wide margin, and the only schedule that works is one they will
actually keep.

## Engagement is the other half

Ten people a week, named. Not "engage more". For each: who they are, why they
matter to this user, and what kind of comment lands with them. Mix:

- 3 who are visibly above the user in reach
- 4 peers at roughly the same level
- 3 in the audience the user sells to

Twenty minutes a day, before posting, not after. Comments on other people's
posts do more for a small account than the posts themselves.

## Write the plan to disk

Write `~/.claude/bylinekit/plan.md` with the week, each slot carrying the idea
and the specific ledger entry it draws on. Then append the themes used to
`log.md`, so next week's plan can avoid them.

Format each slot so `/bl-draft` can pick it up without a conversation:

```
## Wednesday - the proof
idea: proposal time cut from 4h to 20min
ledger: "Cut proposal time ~4h -> ~20min with saved template, Q1 2025"
opener candidates: Interval, Delta
```

## Rules

- **Never fill a slot with nothing.** If there is no material for Thursday, say
  so and drop the slot. Three real posts beat four with a filler.
- **One theme per week, three angles.** Repetition across a week is how people
  come to know the user for something.
- **Every slot names its ledger entry.** A slot with no entry is a slot that
  will produce `{{placeholders}}` on Wednesday.