Skip to main content
These prompts are ready to copy, paste, and adapt: swap in your own channel names, services, and repositories. Each comes with the reason it works, so you can keep the mechanism when you change the words.

​
First messages in a new channel

Ask what Claude can access from this channel:
Why it works: what Claude can do differs per channel, and without this grounding it may suggest tasks it can’t do here. Get a personalized starting point:
Why it works: it’s a discovery task with a bounded output. “Three tasks” gives you a list you can judge in ten seconds, and the answer doubles as a menu of next tasks. Start with a low-stakes task:
Why it works: “since Monday” bounds the task, and you can grade the result yourself because you were there.

​
Shape how the channel works

The prompts below set channel-wide behavior that applies to every thread, not just yours.
Why it works: “remember for this channel” is the explicit save instruction; preferences mentioned in passing usually aren’t kept.
Why it works: it’s stated as standing channel behavior, so it persists instead of lasting one conversation.

​
Check and correct memory

Why it works: memory is a curated note and Claude decides what’s worth keeping, so checking is the only way to know what stuck.
Why it works: it names the specific entry. “Clean up your memory” makes Claude guess what’s stale; naming the entry doesn’t.
Why it works: a correction in a thread fixes that thread only. This sentence is what turns it into a standing fix that applies to everyone’s future threads.

​
Manage routines

Create, audit, and stop the scheduled jobs Claude runs in this channel. For paste-ready schedules by scenario, like a daily standup summary or a weekly digest, see Routine recipes.
Why it works: the schedule is specific, and the post’s contents are named, so every rollup has the same shape and is comparable to last week’s. “Anything still unrouted” turns a recap into a sweep: nothing gets dropped silently.
Why it works: schedules are channel state, and someone else may have set them up. Auditing first beats creating a duplicate digest.
Why it works: it names which job. Any channel member can disable a scheduled job; you don’t need to find an admin to stop a noisy routine.

​
Steer work mid-thread

Reply in the same thread; once Claude is working there, you don’t need to @-mention it again.
Why it works: the reply lands in the session that’s running the task, with full context. Asking in a new thread instead queues a second session that knows nothing about the first.
Why it works: redirecting in-thread keeps everything the session has already learned. “Post the diff here before applying” adds a checkpoint, so the channel reviews the change before it lands.
Why it works: the thread is durable but the isolated workspace behind it isn’t. Work posted to the thread or pushed to a branch survives idle recycling; files that exist only in that workspace don’t.

​
Task starters, by shape

Each entry in the use case library gets one starter here; the linked page has the full setup and the reasoning behind its prompts.