swarm-advanced

name: swarm-advanced description: Operate advanced swarms with adaptive routing, health-aware scaling, and validated convergence patterns. allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Task, TodoWrite model: sonnet x-version: 3.2.0 x-category: orchestration x-vcl-compliance: v3.2.0 x-cognitive-frames:

• HON
• MOR
• COM
• CLS
• EVD
• ASP
• SPC

STANDARD OPERATING PROCEDURE

Purpose

Manage sophisticated swarm behaviors—adaptive topologies, quorum decisions, failover, and convergence—while keeping evidence and confidence ceilings explicit.

Trigger Conditions

• Positive: high-scale swarms, adaptive routing, quorum/consensus needs, burst control, health-based scaling.
• Negative: small fixed teams, prompt-only edits (route to prompt-architect), or new skill weaving (route to skill-forge).

Guardrails

• Skill-Forge structure-first: keep SKILL.md, examples/, tests/ current; add resources//references/ or log remediation.
• Prompt-Architect hygiene: capture HARD/SOFT/INFERRED constraints (scale, latency, resilience), use English-only outputs, and provide ceilinged confidence.
• Swarm safety: enforce registry, health probes, circuit breakers, and rollback; maintain hook latency budgets.
• Adversarial validation: test churn, partition, rate limits, and quorum failure; capture evidence and metrics.
• MCP tagging: log runs under WHO=swarm-advanced-{session} and WHY=skill-execution.

Execution Playbook

1. Intent & constraints: define mission, scale, and SLOs; confirm inferred needs.
2. Topology & health: choose topology, set health probes, and autoscaling rules.
3. Routing & delegation: set sharding, retries, backoff, and escalation paths.
4. Safety nets: plan failover, rollback, and isolation; guard shared state.
5. Validation loop: run churn/partition drills, measure convergence and latency, and store evidence.
6. Delivery: provide topology, evidence, risks, and confidence ceiling.

Output Format

• Swarm objective, constraints, and topology.
• Health model, routing rules, rollback/failover plan.
• Validation evidence (churn, partition, quorum) and risks.
• Confidence: X.XX (ceiling: TYPE Y.YY) - rationale.

Validation Checklist

• Structure-first assets present or ticketed; examples/tests aligned to advanced swarm cases.
• Registry and health probes verified; rollback/failover documented; hook budgets met.
• Adversarial/COV runs stored with MCP tags; confidence ceiling declared; English-only output.

Completion Definition

Swarm is stable when health and convergence targets are met, failover is validated, risks are owned, and evidence is persisted with MCP tags.

Confidence: 0.70 (ceiling: inference 0.70) - Advanced swarm SOP rebuilt with skill-forge scaffolding and prompt-architect evidence/confidence guardrails.