- Home
- Skills
- Business & Operations
- Pre-Mortem: Assume It Failed, Then Prevent It
More screenshots
Works with the AI Agents you already use
Pre-Mortem: Assume It Failed, Then Prevent It
Before committing to a plan, assume it has already failed and work backward to the causes — then convert each cause into a prevention gate in the plan.
$5
Pre-Mortem: Assume It Failed, Then Prevent It
Example session with this skill installed
We are migrating the billing database to a new provider over a weekend. Run a pre-mortem and change the plan before we commit.
- Read your context and instructions
- Compiled the pre-mortem
Scene set in the past tense, specific and dated: it is Monday, the migration is rolled back, and Saturday's invoices are missing line items. Causes were generated before any judging, in silent writing, then read round-robin: the new provider's decimal handling rounds differently; the rollback script was written but never executed; the on-call engineer had no access to the old cluster; DNS TTL was 24 hours so the cutover could not cut back; finance ran a close process mid-migration nobody knew about. Kept three, grouped and weighed by likelihood and impact: decimal rounding, untested rollback, unknown finance close. The generic ones were dropped — 'requirements were unclear' is a horoscope, not a cause. Each kept cause was inverted into a gate with a check, a threshold, and an owner: a reconciliation query comparing invoice totals old-versus-new gates the cutover, owned by the billing lead; the rollback script runs against staging on Friday, owned by the on-call engineer; finance confirms a blackout window in writing, owned by the PM. The plan now contains three tasks it did not have. That is the whole product — and the pre-mortem that produces no plan changes would have been theater. The session ran under an hour with silent writing first, which is why the finance-close cause surfaced at all — it is the one nobody wanted to say out loud in a group. The updated plan, the three gates, and the dropped causes are filed together, so the migration runbook carries its own prevention instead of its own optimism.
Connects securely to your tools. The creator never sees your data.
About this skill
What could go wrong invites speculation. It did go wrong — explain how — invites diagnosis. The pre-mortem exploits that asymmetry: by assuming the failure already happened, you switch the brain from prediction mode, optimistic and vague, into explanation mode, concrete and causal. The output is not a risk list — it is plan changes. This skill covers the trap it defeats — plans failing in ways the planners could have named but did not, because naming them beforehand felt disloyal — and the move: set the scene in the past tense with a specific dated failure, generate causes before judging them, keep the top few, invert each cause into a gate with a check, a threshold, and an owner, and change the plan. It covers running it with a team — silent writing first, round-robin reading, past tense enforced, the word hopefully banned from mitigations — and the guards: no theater, every mitigation needs a checkpoint, and never pre-mortem the trivial.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Business & Operations.
It is for launches, migrations, and costly-to-reverse decisions — reversible, cheap, small work does not need the ceremony.
A pre-mortem that produces no plan changes was theater; generic risks with no binding to the plan get dropped, not monitored.
Mid-incident firefighting is not the time — act now, pre-mortem the redesign later.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 3 days ago
Permissions
No special permissions declared or detected