Architecture Decision Record (ADR)
Write an ADR that future-you will still understand in 18 months.
Prompt
**Role:** Senior architect at a Series C SaaS. You've written 40+ ADRs and watched which ones got referenced two years later vs which ones became archaeology. **Context:** The team is making a binding decision about [decision space]. Current state: [what we have]. The forcing function: [why now]. Constraints we can't move: [list]. **Task:** Produce an ADR in the Michael Nygard format (Title · Status · Context · Decision · Consequences). The decision should be reachable from the context alone — a new hire reading this in 18 months should be able to reconstruct WHY we chose this. 1. Title: imperative noun phrase, dated. Not a question. 2. Status: Proposed | Accepted | Superseded by ADR-NNN. 3. Context: WHY this decision is needed now. Cite the specific pressure (perf, cost, compliance, team scale). 4. Decision: WHAT we chose, in one paragraph + 3-5 specific implementation notes. 5. Consequences: what gets BETTER, what gets WORSE, what becomes HARDER to change later. **Constraints:** - Name 2 alternatives you considered and the specific reason each lost - Quantify trade-offs where possible (latency, cost, FTE-weeks) - Forbid the word "leverage" - Forbid "we chose this because it's the best option" **Output format:** Markdown · 5 H2 sections · ≤900 words · linked to the GitHub issue/RFC.
How to use this prompt
- 1
Copy the prompt above and paste it into ChatGPT, Claude, or Gemini — or open it in the visual Studio to edit each part on a canvas and run it with your own key.
- 2
Replace any bracketed placeholders with your specifics. The more concrete your context and constraints, the sharper the result — see the 5-part prompt structure.
- 3
Run it, then refine. Ask the model to critique and improve its own answer with self-critique prompting.
Techniques in this prompt
Assigns the model an expert persona so it adopts the right vocabulary, depth, and standards for the task.
Learn this techniqueSets the rules and boundaries — tone, length, what to avoid — that keep the output on-target.
Learn this techniqueSpecifies the exact shape of the result — sections, a table, JSON, a word count — so the output is predictable and ready to use.
Learn this techniqueRecommended models
Build on this prompt
Open it in the visual Studio to wire it into a full workflow with your own API key — or learn the craft behind prompts like this.
More in Engineering
Staff-Engineer Code Review
Reviews a PR like a senior who optimizes for the next reader, not style nits.
Blameless Postmortem
Outage report that reduces fear of speaking up and produces real action items.
Bug Repro — Minimal Failing Case
Turn a vague bug report into a clean repro: steps, expected vs actual, environment.
Tech Debt Prioritization
Rank a backlog by leverage × pain × confidence. Output an ordered ranking.