This is STM / swarm security Phase 0 — PeerObservation, StateTransitionManager, threat T1–T7, mesh evidence. Not coordination Phase 0F (CLI freeze). Disambiguation:
docs/coordination/PHASE-0-TERMINOLOGY-DISAMBIGUATION.md· Coordination hub:docs/coordination/README.md
This folder is the Phase 0 knowledge graph for the Perpetua-Tools swarm and StateTransitionManager work. It is not one plan repeated many times. It is a layered record of how the system moved from P2P security research, to TDD contracts, to STM reconciliation, to multi-agent review, to the v1-to-v2 security and coordination handoff.
Read this page first, then use the LLM-wiki pages under wiki/ for
file nodes, concept nodes, and explicit edges.
Penultimate pre-v2 checklist (every plan disposition):
PHASE-0-MASTER-PLAN-2026-07-27.md
The root problem behind these documents is that useful peer-orchestration ideas existed as separate modules, tests, and review findings, but the product needed a single decision boundary. Phase 0 discovered that append-only peer observations, confidence scoring, liveness checks, replay protection, witness quorum, and threat-model controls only become operationally meaningful when they converge into a state-transition owner with bounded queues, explicit failure semantics, and production wiring.
In first-principles terms:
- A peer report is evidence, not state.
- State changes need a monotonic decision boundary.
- Liveness and confidence are coupled; neither is sufficient alone.
- BFT-style patterns must be sized to the actual deployment premise.
- Multi-agent work needs source-of-truth discipline, or review findings become branch drift and duplicated fixes.
| Order | Start here | Why |
|---|---|---|
| 1 | PATTERN-SYNTHESIS.md |
Extracts the reusable P2P security patterns that seeded Phase 0. |
| 2 | DELIVERABLE-1-PEER-OBSERVATION-MODEL-REGENERATED-ITERATION-2.md |
Canonical repaired PeerObservation and confidence model. |
| 3 | DELIVERABLE-2-HEARTBEAT-LIVENESS-REGENERATED.md |
Liveness, timeout hierarchy, and hysteresis expectations. |
| 4 | DELIVERABLE-4-THREAT-MODEL-REGENERATED.md |
Threat IDs and adversarial controls used by later plans. |
| 5 | PHASE-0-FIX-3-AND-MEDIUM-ITEMS-DECISION-BRIEF.md |
Reconciles STM model conflicts and medium-risk decisions. |
| 6 | 2026-07-11-state-transition-manager-integration-plan.md |
Converts separate evidence modules into one STM pipeline. |
| 7 | README-PR203-BLEND.md |
Navigation hub for the PR #203 blend and next-agent handoff. |
| 8 | 2026-07-12-autoplan-final-approval-gate.md |
Final approval gate for the Phase 2 STM concurrency/dedup increment. |
| 9 | 2026-07-12-stm-next-increment-plan.md |
Closed plan for threat-model re-check, production wiring, and hardening. |
flowchart TD
Patterns[P2P security patterns] --> PeerObservation[PeerObservation schema]
Patterns --> ThreatModel[Threat model T1-T7]
Patterns --> SwarmSecurity[Swarm security analysis]
PeerObservation --> Confidence[Confidence scoring]
PeerObservation --> TDD[TDD fixtures and vectors]
Confidence --> STM[StateTransitionManager]
Heartbeat[Heartbeat and liveness] --> STM
ThreatModel --> STM
Medium[M1-M7 decision matrices] --> DecisionBrief[Decision brief]
DecisionBrief --> STM
TDD --> Phase0TaskList[Phase 0 task list]
Phase0TaskList --> Phase1Scope[Phase 1 scope]
STM --> PR203[PR #203 blend]
PR203 --> Phase2Blockers[Phase 2 blockers]
Phase2Blockers --> ApprovalGate[Final approval gate]
ApprovalGate --> NextIncrement[STM next increment]
SwarmSecurity --> SecurityPolicy[SECURITY.md]
NextIncrement --> DocsNext[docs/next handoff]
| Page | Use it for |
|---|---|
wiki/files.md |
File-by-file node inventory and downstream role. |
wiki/concepts.md |
Concept dictionary with primary source files. |
wiki/edges.md |
Explicit source-to-target node relationships. |
wiki/security-trace.md |
Threat, liveness, dedup, and policy links into security docs. |
- Forward-looking PT work:
../next/README.md - PT security policy:
../../SECURITY.md - Orama companion security policy:
orama-system/SECURITY.md - Orama v2 security foundation:
docs/v2/31-security-harness-excellence-plan.md,docs/v2/32-agentic-security-controls.md,docs/v2/39-maestro-owasp-genai-reference.md