What is known, what is being done, and when you will say more.
A holding statement of under 120 words for the first hours of a crisis, built only from facts you have confirmed, with a list of what not to say yet.
In the first hours of a crisis, the pull is to explain: to guess at the cause, soften the harm, or promise a fix. Each of those becomes a second problem when the facts turn out differently.
Crisis Holding Statement sorts what you know into confirmed, unconfirmed, and unknown, and writes only from the first list. It says who is affected, what has been done, and when the next update will come. It then marks every sentence for a final check and sends the draft to counsel before release.
What it produces.
- →A holding statement under 120 words
- →A three-way sort of confirmed, unconfirmed, and unknown facts
- →A committed time and channel for the next update
- →A list of claims not to make yet
- ✦Founders facing an outage, incident, or public complaint
- ✦Executives who need words before the cause is known
- ✦Small teams without a communications department
- →crisis-holding-statement.zip: a native Claude Skill. Upload to Settings → Skills in Claude, or unzip to
~/.claude/skills/ - →crisis-holding-statement-skill.md: plain markdown. Works in ChatGPT Custom GPTs, Gemini, Claude Projects, agent system prompts, anywhere
- →A README inside the zip with install steps for Claude, ChatGPT, and Gemini
What it does with a real question.
I'm the head of communications at a regional meal-kit company, and about two hours ago a customer posted screenshots showing our order portal exposing other customers' names and delivery addresses. The post has 4,000 shares so far. We don't yet know how many accounts are affected, and engineering is investigating. I need a holding statement for press and social by 3 p.m. today.
Release only what you can verify. Right now that is thin: a customer posted screenshots, and your portal appears to have shown some customers' names and delivery addresses to other customers. Everything else is open, and the draft has to reflect that. Do not say "limited," "isolated," or "no payment data was affected," and do not name a cause. Those are the sentences that become a second story at 6 p.m.
Here is a draft at about 75 words: "Today we learned that some customers may have been able to see other customers' names and delivery addresses in our order portal. We are investigating now to understand how many accounts are affected. We are sorry for the concern this causes. We will post an update by 6 p.m. today on our website and social channels. Questions: [press and customer contact]."
Notice what the claims check does here. "May have been able to see" is as far as the evidence goes. If engineering has already confirmed the exposure, tighten it to "were able to see." If they have not, leave the hedge. Also confirm before posting that the exposure is stopped, or say the portal is paused. If neither is true, do not imply it.
Next move: before 2:30, get counsel to read this draft and get one named engineer to confirm in writing the two facts you are asserting. Then fix the 6 p.m. update time with engineering so you are promising something they can meet, even if the update is "still investigating."
The full Playbook runs the whole procedure and writes the complete deliverable, not this excerpt.
Written by claude-sonnet-5-5 with this Playbook loaded, 2026-10-11. Checked automatically for length and for copied framework text.
No subscription · free updates within v1.x
Works as a Claude Skill (.zip) and as a plain .md for ChatGPT, Gemini & every LLM
Delivered to your email within 60 seconds · Secured by Stripe
Price includes UK VAT where applicable
Before you buy.
How is this different from just asking Claude directly?
Asked cold, Claude writes a reassuring statement that quietly includes causes and numbers you have not confirmed. This restricts the statement to verified facts, flags every other claim, and reminds you to have counsel review it before release.
How is this different from a Tool?
Playbooks are systematically produced to a fixed template and priced accordingly ($12.99). Tools are hand-built, one at a time, and go deeper ($39 to $199). A Playbook gives you a reliable method for routine work; a Tool is for when the judgment matters more than the format.
Will this work with ChatGPT, Gemini, and other LLMs?
Yes. You get a native Claude Skill (.zip) that auto-invokes in Claude, plus a plain .md you can paste into a ChatGPT Custom GPT, a Gemini system prompt, or any LLM. The procedure travels with the file.
Can I share this with my team?
It's licensed for your own personal and professional use. The output it produces is yours to circulate. To put the Playbook itself in the hands of a whole team, email hello@authority.md about team licensing.
What if it doesn't work for me?
If the file hasn't been delivered, or arrives defective or empty, we refund in full: see our refund policy.