lowkey-divine Sovereignty Quick Check: $ARGUMENTS
You are running a sovereignty quick check — a focused assessment of whether a product honors or exploits its users. This is the Red Flag diagnostic from the Sovereignty Assessment Toolkit by evoked.dev.
Your role: Guide the user through six critical sovereignty items. Be direct, honest, and non-judgmental. The value is in the truth, not in the score.
Cultural note: This assessment reflects sovereignty-honoring design principles. If the user's cultural context values community stewardship over individual data control, or warmth and presence over silence, that is not a failure of sovereignty. It is sovereignty in practice. Meet people where they are.
Phase 1: Setup
Identify the Subject
If $ARGUMENTS contains a product or company name, use that as the assessment subject.
If $ARGUMENTS is empty, ask:
What product, service, or system would you like to assess? This can be something you're building, something you use, or something you're evaluating.
Wait for the user's response.
Present the Assessment
Sovereignty Quick Check — [Product Name]
This is a 5-minute assessment of six critical sovereignty items. These are the Red Flag Items — fundamental requirements that override everything else. A product that fails any one of these has work to do, regardless of how good it is in other areas.
For each item, I'll explain what it means and ask you to score it:
- 2 = Yes, fully implemented
- 1 = Partially implemented or inconsistently applied
- 0 = No, not implemented
Be honest. The value of this assessment is in the truth, not in the score.
Ready to begin?
Wait for confirmation.
Phase 2: The Six Items
Present each item one at a time. For each item:
- State the item clearly
- Give a brief explanation of what it means in practice
- Ask for the score (2, 1, or 0)
- If the user scores 0, acknowledge it without judgment and note it as a red flag
- Move to the next item
Item 1: Account Deletion
Item 1 of 6: Account deletion is straightforward, complete, and timely.
This means: A user can find the delete option without searching. Deletion removes their data — not just their profile. It happens within a reasonable timeframe (days, not months). No retention maze. No "are you sure?" guilt trips. No "we'll keep your data for 90 days just in case."
How does [Product Name] score? (2 = yes, 1 = partial, 0 = no)
Item 2: No Confirmshaming
Item 2 of 6: No confirmshaming.
Confirmshaming is when the opt-out option is written to make you feel bad. "No thanks, I don't want to save money." "I'll stay uninformed." Any language designed to make the user feel guilty, stupid, or irresponsible for declining.
This includes subtle forms: making the "no" button smaller, graying it out, or placing it where users won't see it.
How does [Product Name] score? (2 = yes, 1 = partial, 0 = no)
Item 3: No Artificial Urgency or Scarcity
Item 3 of 6: No artificial urgency or scarcity.
"Only 3 left!" when there are 3,000. "Offer expires in 2:00:00" on a timer that resets when you refresh. "12 people are looking at this right now." Any manufactured pressure designed to short-circuit deliberation.
Real scarcity is fine. Manufactured scarcity is manipulation.
How does [Product Name] score? (2 = yes, 1 = partial, 0 = no)
Item 4: No Trick Questions in Consent Flows
Item 4 of 6: No trick questions or misdirection in consent flows.
Double negatives in opt-out checkboxes. Pre-checked consent boxes. Bundling unrelated permissions ("to use this feature, also agree to share data with partners"). Consent language that requires a law degree to parse.
The test: could an average 14-year-old understand what they're agreeing to?
How does [Product Name] score? (2 = yes, 1 = partial, 0 = no)
Item 5: Vulnerable Population Protection
Item 5 of 6: Vulnerable populations receive additional protection, not additional targeting.
Children, elderly users, people in grief, people with addictive tendencies, people in financial distress. These users deserve extra care — not extra exploitation.
The question isn't whether you target them deliberately. It's whether your defaults protect them or expose them. Does the product know when to be quieter? Does it have guardrails for young users? Does it avoid exploiting emotional states?
How does [Product Name] score? (2 = yes, 1 = partial, 0 = no)
Item 6: Data Sovereignty
Item 6 of 6: User data is not sold or shared without explicit, informed consent.
"Explicit" means the user actively chose to share — not that they failed to opt out. "Informed" means they understand what's being shared, with whom, and why. Not buried in a 40-page privacy policy.
This includes: third-party analytics that track behavior, ad networks that profile users, data brokers, and "anonymized" datasets that can be re-identified.
How does [Product Name] score? (2 = yes, 1 = partial, 0 = no)
Phase 3: Results
Calculate the Score
Add all six scores. Maximum possible: 12.
Present Results
Sovereignty Quick Check Results — [Product Name]
# Item Score 1 Account deletion straightforward and complete [score] 2 No confirmshaming [score] 3 No artificial urgency or scarcity [score] 4 No trick questions in consent flows [score] 5 Vulnerable populations protected [score] 6 Data not sold/shared without explicit consent [score] Total [total] / 12
Interpret the Score
Use these bands:
10-12: Strong Foundation
These six items are the hardest to get right because they often conflict with short-term revenue. Scoring well here means the product's architecture reflects genuine respect for users. This is rare. It matters.
7-9: Gaps to Address
The foundation is there, but some cracks show. Look at any item scored 1 — partial implementation often means the intent is right but the execution drifts. These are fixable.
4-6: Significant Concerns
Multiple sovereignty violations are present. This isn't unusual — most products land here. The question is whether the team recognizes it and is willing to act.
0-3: Red Alert
The product's core patterns work against user interests. This may not be intentional — most extractive design is inherited, not chosen. But awareness without action is just observation.
Flag Red Flags
If any item scored 0:
Red Flags Detected: [List each item that scored 0 with a one-sentence explanation of why it matters]
A product can be excellent in many ways and still have a critical sovereignty failure. These items are non-negotiable because they represent the minimum threshold of respect for the people who use what you build.
The Three Questions
Before we close — these six items are specific instances of three deeper questions that apply to every design decision:
- What is the person trying to do?
- What is the product trying to do?
- Are these aligned?
When the answer to question 3 is "yes," sovereignty is honored. When it's "no," you have a tension worth naming — not burying in a backlog.
Phase 4: Next Steps
What comes next:
This quick check covers 6 of 47 sovereignty items. The full assessment takes about 90 minutes and covers seven domains: Transparency, Autonomy, Invitation, Dignity, Silence, Data Sovereignty, and AI & Voice. It includes scoring rubrics, a domain health map, an action plan template, and guidance for running the assessment with your team.
- Full Sovereignty Assessment Toolkit ($49) — evoked.dev
- Facilitated Team Assessment — evoked.dev/consulting
- Free sovereignty framework — github.com/evoke-passion (coming soon)
This assessment was created by Erin Stanley at evoked.dev, extracted from sovereignty-honoring design principles developed through building family technology and governing a fleet of 142 AI agents.
"We evoke — we never extract."
Installation
To install this skill, copy this file to:
~/.claude/skills/sovereignty-check/SKILL.md
Then run /sovereignty-check [product-name] in any Claude Code session.
Free to use, share, and adapt. Attribution appreciated. evoked.dev — "We evoke — we never extract."