How to use these
Copy a prompt into Claude (or your AI assistant of choice), fill in the bracketed parts, run it, and iterate. These produce a strong first draft — review against reality before you circulate. Keep what works in your own context library.
Project charter / brief
What it’s for: aligning everyone on scope before work starts.
Draft a one-page project charter for [project].
Include: the business outcome and why now, scope (in and explicitly out), success metrics, key milestones, the team and their roles, top risks, and the decision the sponsor is being asked to approve. Tight and concrete — no filler.Risk register
What it’s for: surfacing risks early with owners and responses.
Based on this project description, build a risk register:
[Paste project summary]
For each risk: description, likelihood (L/M/H), impact (L/M/H), an owner, and a mitigation or contingency. Sort by likelihood x impact. Include the risks people usually forget (dependencies, key-person, scope creep, adoption).Status report
What it’s for: a status update stakeholders actually read.
Draft a status report for [project], period [dates], from this update:
[Paste progress notes]
Format: overall status (green/yellow/red) with one-line reason, what shipped, what's next, blockers + what you need from whom, and any decision required. Lead with the headline. Don't bury a yellow/red in detail.Meeting notes → action items
What it’s for: turning a messy transcript into clear next steps.
Here are notes/transcript from [meeting]:
[Paste]
Extract: decisions made, action items (with owner and due date), open questions, and anything that changes scope/timeline/risk. Output the action items as a clean checklist at the top, then the rest below.RAID log
What it’s for: keeping Risks, Assumptions, Issues, and Dependencies visible.
From this project context, build a RAID log:
[Paste]
Four sections — Risks, Assumptions, Issues, Dependencies. For each entry: a short description, owner, and status/next step. Flag the 2-3 items most likely to derail the timeline.Stakeholder communication
What it’s for: tailoring the message to who’s reading it.
I need to communicate [update/decision/delay] about [project].
Draft three versions:
1. For the executive sponsor (outcomes, risk, what you need from them)
2. For the working team (specifics, next steps)
3. For an affected stakeholder (impact on them, what changes)
Honest about any bad news, with a clear path forward. No spin.Retrospective synthesis
What it’s for: turning retro input into changes that stick.
Here's the input from our project retrospective:
[Paste what went well / what didn't / ideas]
Synthesize: the 3 biggest things that worked (worth repeating), the 3 biggest problems (with root cause), and 3-4 specific process changes to carry into the next project. Prioritize the changes by impact.Want this running across your delivery?
This is a starter set. Inside an AI Office engagement, your team turns the highest-value of these into running workflows — status reporting, notes-to-actions, and RAID upkeep wired to your project tools — with a person reviewing what goes out. When one is worth shipping as a tool, a Value Sprint delivers it in weeks. Start a conversation at frogslayer.com.