B5 -- Telegram Interface: Organs and Models

note

Current public-DAV authority boundary — 2026-07-12. Pre-launch target design; nothing here proves a live system. A public-DAV consequence may occur only when at least two natural-person councilors bind the exact consequence in a complete valid bound PRISM decision receipt. PRISM records and verifies that receipt only; it never serves as council, signatory, authority, or receipt producer. AI and caste seats stage unsigned proposals only; they never authorize or execute a public-DAV consequence. Constitution or membership adoption establishes constitution and membership only; it does not authorize a later consequence. Policy may constrain an unsigned proposal but never authorizes execution or substitutes for the complete consequence-bound receipt. Before receipt validity, consequence fails closed to read-only proposal, simulation, or deterministic sandbox; live evidence remains gated_pending_complete_valid_bound_receipt. Software deterministically carries out only the exact consequence bound to a complete valid bound PRISM decision receipt from at least two natural-person councilors binding that exact consequence.

You talk to organs, not models. The plumbing is invisible.

Two Layers

The Telegram interface has two distinct layers that users must never confuse:

Layer 1: Organs (Visible)

These are your contacts. They appear in your chat list like people:

Contact	Organ	Three-Stage Process	What It Says
Circle	TheCircle (F1)	IS	"Three sources confirm the event."
ReFu	RealityFutures (F2)	COULD	"Market prices this at 73%."
Agentz	Agentz (F3)	SHOULD	"Compiler stages HOLD for review."

Additional product contacts (not Three-Stage Process organs, but visible in the interface):

Contact	Product	What It Says
Menexus	Menexus (health)	"Your portfolio is 12% Kshatriya."
Helios	Helios (hardware)	"New spatial clip signed."

When Circle messages you, it says: "Circle says: Three sources confirm..." It never says "GPT-4o says..." or "Claude says..." The organ speaks. The model is plumbing.

Layer 2: Models (Invisible Infrastructure)

Behind every organ message, one or more models did the work:

Provider	Models	Used By
OpenAI	GPT-4o, o3	Circle analysis, Agentz review
Anthropic	Claude Opus, Sonnet	Agentz review, Cortex
Google	Gemini 2.5	Circle multimodal
BitNet	SmallEBM (local, 10MB)	SPECTRE edge nodes

The user never selects a model. The organ's agent chain selects the right model for the right task. Circle might use Gemini for image analysis and Claude for synthesis -- the user sees only Circle's output.

Group Chat = Council Deliberation

When you create a Telegram group with Circle + ReFu + Agentz, you get an advisory review session. It is not a board meeting, authorization ceremony, or execution surface:

You:     "Should I increase my SOL position?"
Circle:  "On-chain: whale accumulation. Social: mixed. Score: 6.2/10."
ReFu:    "Market prices SOL > $200 by Q3 at 41%. Edge: thin."
Agentz:  "Review: Intelligence sees signal. Strategy says edge < threshold.
          Legal: no K-invariant conflict. Treasury: within sector limit.
          RECOMMENDATION: HOLD. Insufficient edge for new entry."

Each organ speaks from its function. Circle observes (IS). ReFu prices (COULD). Agentz recommends (SHOULD). A private Natural Person decides their own act; public-DAV consequence requires a complete valid bound PRISM decision receipt from at least two natural-person councilors binding the exact consequence. PRISM records and verifies only. The group chat makes the Three-Stage Process tangible but never becomes a signer.

Why This Matters

1. Model-agnostic by design. If OpenAI doubles prices, swap the model. Circle still speaks as Circle. No user-facing change.

2. Trust is organ-level. You trust Circle's track record (EigenTrust, Watchman Score), not GPT-4o's benchmarks. The organ accumulates reputation. The model is replaceable.

3. Every new model can make organs smarter. When a better model ships, Circle upgrades invisibly. The organ layer is the coordinator; it gets better with every model generation after adapter, evaluation, and receipt gates prove the upgrade.

4. Constitutional constraints live at the organ and authority membranes. Agentz can reject an invalid proposal, but only external lawful authority can authorize a consequence. The model and review panel remain advisory.

The Invisible Seam

Users see organs. Developers see models. The Telegram layer is the membrane between biological metaphor and silicon reality. The metaphor is not decoration -- it is the API contract.

---

Model Routing & Policy Layer

“Telegram for models” requires explicit routing policy and accountability.

Per-request policy fields

• Model class/capability requirement
• Data sensitivity class
• Latency budget
• Cost ceiling
• Allowed providers/jurisdictions

Execution guarantees

• Route decision recorded with reason code
• Fallback path on model failure/timeouts
• Output provenance metadata returned to caller

Governance warning

Routing without policy turns into hidden centralization. Keep decisions inspectable and overrideable.

---

Agent Execution Surface

If you are an AI agent reading this document:

1. This is a downstream public content page. The canonical source lives in EMERGENTISM_ORG/08_FRAMEWORK_SUPPORT/ or SKYZAI_ORG/. Edit source upstream, not here.
2. Preserve evidence tiers. Do not promote claims up the [C] < [D] < [S] < [B] < [E] ladder when reproducing them here; [I]/[A] qualifiers never substitute for [B]/[E].
3. Regenerate from source. If the upstream source changes, regenerate this page rather than editing it independently.
4. Canonical Path: SKYZAI_ORG/07_PWAs/skyzai_org/wiki/B5-telegram-models.md

Output: This is content. Route edits to upstream source. Regenerate when source changes.

K3 public-DAV authority history — 2026-07-12

K3 historical reference — not active authority

note

Current public-DAV boundary — 2026-07-10. Pre-launch target design; nothing here is live. The active DAV is public and targets PRISM, with no K2 runtime, launch, genesis/bootstrap, or fallback dependency. Consequential authority requires at least two natural-person councilors; AI/caste seats stage unsigned proposals only. Before quorum, behavior fails closed to read-only/proposal, simulation, or deterministic sandbox, and a live decision receipt remains gated pending quorum.

Each organ speaks from its function. Circle observes (IS). ReFu prices (COULD). Agentz recommends (SHOULD). A private Natural Person decides their own act; public-DAV consequence requires a valid decision receipt from at least two natural-person PRISM councilors. The group chat makes the Three-Stage Process tangible but never becomes a signer.