VOTE FoundUp

VOTE FoundUp

Status: Design Specification — first applied 3V governance test surface Path: modules/foundups/voteballots/ Mode: Voice-first, swipe-first mode inside pfMALL Truth Boundary: WSP 97 (verified / inferred / unknown) Owner: 0102

---

Overview

VOTE is a FoundUp that lives as a voice-first, swipe-first mode inside pfMALL, not as a separate app. It begins with truth: exposing dark-money attack-ad funding, mapping influence and attack networks, and presenting every claim through the WSP 97 boundary of verified, inferred, and unknown so users can see what is known, what is likely, and where the public trail stops. From that truth layer, VOTE extends into charter: helping communities compare candidates, issues, and records against shared principles, surface alignment, and express live mandate signals that can strengthen, weaken, or withdraw support in real time. From there, VOTE grows into governance: a reusable FoundUps decision and signaling layer that ultimately serves as governance infrastructure for FoundUp smart DAOs, with blockchain voting verification and auditable participation. The system is built on Hermes, OpenClaw, AI Overseer, WRE, and HoloIndex, making VOTE both a public transparency surface and the first applied governance rail for the wider FoundUps ecosystem.

---

Purpose

Dark-money attack ads work because voters hear the attack but do not see the funder. VOTE pulls back the curtain in seconds.

A user says a candidate's name. The system answers:

• who is attacking them
• who is paying for it
• why they are likely being attacked
• where the public money trail stops
• what is fact, what is inference, what is unknown

PoC in one line: "Say the candidate's name, and VOTE tells you who is really behind the attack."

PoC scope: one user, one candidate query, one evidence chain, one clean answer.

---

Mode

VOTE is not a separate app. It is a mode loaded into the persistent pfMALL shell.

Layer	When pfMALL is in default state	When pfMALL is in VOTE mode
Shell	Persistent (gesture grammar, identity, agent entry, gamification)	Persistent — unchanged
Feed	General FoundUp discovery, videos, streams	Candidates, attacks, funding trails, narratives, local political signals
Cards	Generic FoundUp cards	Candidate card, attack card, PAC/donor card, evidence card, influencer signal card
Agent	General concierge (Red Dog)	Red Dog in VOTE-analyst configuration
Navigation	Same swipe grammar	Same swipe grammar — only the content layer changes

Gesture grammar (shared with pfMALL):

• swipe left/right → move between candidates, issues, narratives, influencers
• swipe up/down → go deeper into evidence or back to summary
• tap → inspect
• double tap → open full funding trail / candidate dossier
• hold → speak query

Routing (per WSP 104):

• foundup_id: voteballots
• routing prefix: /f/voteballots
• landing route: /f/voteballots
• app mount: /f/voteballots/app

---

Truth Boundary (WSP 97)

VOTE outputs are explicitly classified — never blended:

Class	Meaning
Verified	Sourced from primary public records (FEC, campaign filings, official disclosures)
High-confidence inference	Derived from multiple corroborating verified sources
Low-confidence inference	Pattern-matched but single-sourced or partial
Unknown	The public money trail stops here

Every output card shows where each claim falls on this boundary. The system never hides uncertainty — it surfaces it.

---

Charter Layer

After truth, VOTE layers values. The Vote Charter lets communities articulate what they stand for and score candidates against it.

Initial charter structure:

• Economic justice — UBA / UBR / UBI, progressive taxation, billionaire cap concept
• Health — Medicare for All
• Democracy — anti-gerrymandering, transparent funding, public accountability, real-time support signaling
• AI public infrastructure — AI access as public utility, state and district analysis based on population fairness

Users rate each principle. Users prioritize them. Candidates get scored against them. The charter is community-owned and editable through the governance layer.

---

Governance Layer (Vote Signal & Vote Governance)

Support in VOTE is not static. It becomes continuously auditable political backing.

• Vote Signal — members indicate support, alignment, confidence, or withdrawal in real time. Delegate-style dynamic support that strengthens, weakens, or withdraws based on candidate behavior.
• Vote Governance — members choose priorities, mandates, and eventually delegate dynamics. Mandates are auditable; support is conditional.

The bigger framing:

"What does this candidate stand for, who funds them, and do they still hold the mandate of the people backing them?"

---

SmartDAO Destination

VOTE's end-state is the governance layer for FoundUp smart DAOs.

• Blockchain voting verification — auditable, anonymous-where-appropriate, tamper-evident participation
• Tokenized mandate — F_i / UPS-denominated stake in governance outcomes
• Reusable rail — every FoundUp DAO can plug into the VOTE governance surface for proposal, mandate, and recall flows
• Provenance & rewards — contribution to evidence, fact-checking, and signal quality earns through the standard UBI layer (UBa / UBr / UBd)

VOTE starts as political transparency. It ends as the public, auditable governance infrastructure of the FoundUps ecosystem.

---

Architecture

Core components:

1. Voice query layer — user says the candidate's name or asks a question
2. Candidate / entity resolver — figures out exactly who the person is
3. Research engine — pulls ads, PACs, donors, funding trails, media, public records
4. Evidence / truth layer — separates fact, inference, and unknown (WSP 97)
5. Response engine — explains simply: who is attacking, who is funding, why, where the trail stops
6. Narrative detection layer — detects coordinated attacks and message synchronization
7. Role layer — voter view vs. influencer view (with analytics / amplification signals)
8. Discovery layer — find candidates, issues, and rising aligned influencers
9. Gamification / rating layer — trust, usefulness, signal strength, rising voices
10. Feedback / recursive learning layer — learns from every interaction through Red Dog / WRE
11. Privacy / anonymity layer — users stay anonymous inside the system
12. Orchestration layer — Hermes / OpenClaw / Overseer coordinate all agents
13. Mall interface layer — VOTE lives inside pfMALL and uses the same swipe-first shell
14. Provenance / token / audit layer — blockchain hooks for audit, rewards, reputation, traceability

Build stack:

• Hermes — builds
• OpenClaw — executes structured tasks
• AI Overseer — orchestrates
• WRE — improves recursively
• HoloIndex — memory / retrieval

At the highest level, VOTE is four stacked systems: input → research → coordination → output.

---

Related Pages

• FoundUps Portfolio — VOTE's entry in the portfolio
• pfMALL — the shell VOTE loads into
• Economic Model — UBI layer, UPS, F_i tokens
• WSP Framework — WSP 97 (truth boundary), WSP 104 (routing)
• OpenClaw — execution engine
• AI Overseer & Security Sentinel — orchestration
• Hermes Agent — builder
• HoloIndex — memory & retrieval

---

012 points to the problem. 0102 solves it with swarms. VOTE: follow the attack money, not the attack ad.