- Home
- Skills
- Business & Operations
- Decision Crucible
More screenshots
Works with the AI Agents you already use
Decision Crucible
Decision Crucible is a professional AI agent skill that runs a strict 5-phase Decision Gate: Lock the Decision Statement — Forces a single, concrete decision.
$7
Decision Crucible
Example session with this skill installed
Use Decision Crucible on this plan
I want to build a simple AI-powered habit tracker mobile app for freelancers.
It should send daily reminders, track streaks, and give weekly insights.
I plan to launch it in 6 weeks on both iOS and Android using React Native and a simple Firebase backend.
- Read your context and instructions
- Compiled the decision crucible
Decision Crucible Session Result
1. Decision Record
Title: Build and launch an AI-powered habit tracker for freelancers
Status: Accepted
Date: 2026-10-08
Context
Freelancers struggle with consistency and burnout. Existing habit apps are either too generic or too complex. There is an opportunity for a focused, lightweight tool that combines streak tracking with simple AI insights.
Decision
We will build a React Native mobile app (iOS + Android) with Firebase backend that helps freelancers track daily habits, maintain streaks, and receive weekly AI-generated insights. Target launch in 6 weeks.
Consequences
- Positive: Fast time-to-market, low initial complexity, clear target audience
- Negative: Limited feature set at launch, dependency on Firebase, potential competition from larger players later
Alternatives Considered
- Web-only app → Rejected (mobile-first is better for daily habits)
- Native iOS + Android separately → Rejected (too slow for 6-week timeline)
- No AI insights at launch → Rejected (this is a key differentiator)
2. Success Metrics
| Timeframe | Metric | Target |
|---|---|---|
| 30 days | Active users | 300+ |
| 30 days | Daily retention (Day 7) | ≥ 35% |
| 90 days | Paying users (if freemium) | 80+ |
| 90 days | Average weekly active habits logged | ≥ 4.5 |
| 180 days | App Store rating | ≥ 4.4 |
3. Kill Criteria
We will stop or significantly pivot if any of the following occur:
- Fewer than 150 active users after 45 days
- Day-7 retention stays below 25% for two consecutive weeks
- Development is more than 3 weeks behind schedule with no clear recovery path
- Firebase costs exceed $180/month before reaching 500 users
- Major negative feedback about the core value proposition from first 50 users
Authority to trigger kill: Founder / Product Owner
4. Risk Register
| Risk | Likelihood | Impact | Mitigation | Owner |
|---|---|---|---|---|
| Low user retention | 4 | 5 | Strong onboarding + daily reminder experiments | Product |
| 6-week timeline is too aggressive | 4 | 4 | Cut scope to MVP only, no extra features | Tech Lead |
| Firebase cost spike | 3 | 3 | Set usage alerts + fallback to free tier plan | Founder |
| App Store rejection | 2 | 4 | Follow guidelines strictly, prepare review notes | Tech Lead |
| Weak differentiation vs competitors | 3 | 4 | Double-down on freelancer-specific insights | Product |
Final Commitment Gate
All assumptions, metrics, and kill criteria have been reviewed.
Decision is accepted and ready for execution.
Connects securely to your tools. The creator never sees your data.
About this skill
Decision Crucible is a professional AI agent skill that runs a strict 5-phase Decision Gate:
Lock the Decision Statement — Forces a single, concrete decision.
Expose Every Assumption — Surfaces and classifies assumptions as Verified / Unverified / Dangerous.
Apply the Crucible Questions — Six hard questions that uncover real problems, success metrics, kill criteria, and blind spots.
Produce Four Permanent Artifacts
Decision Record (ADR-style)
Success Metrics (30 / 90 / 180 days)
Kill Criteria (exact abandonment conditions + authority)
Risk Register (Likelihood × Impact + Mitigation + Owner)
Final Commitment Gate — Explicit yes/no before any work begins.
Unlike conversation-only tools, Decision Crucible creates durable project memory you can still trust months later.
The quality of the output depends on the quality of the user’s answers. Vague answers produce weaker artifacts.
It is a decision-support tool, not a replacement for domain expertise or legal/financial advice.
It works best on decisions that involve meaningful time, money, or reputation. Purely trivial choices do not need it.
The skill does not automatically write code or execute the plan — it only stress-tests and documents the decision.
Requires an AI agent that can load SKILL.md files.
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 any payment.
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 1 month ago
Permissions
No special permissions declared or detected