MirrorNeuron Blueprints

mn-blueprints is a self-contained runnable MirrorNeuron workflow blueprint catalog. Each blueprint folder includes its own manifest, configuration, payloads, README, and user-facing SPEC.md.

Quick Start

List available blueprints:

mn blueprint list

Run a catalog blueprint:

mn run <blueprint_id>

Run a checked-in folder directly:

cd <blueprint_id>
mn run --folder .

Run repository tests:

python3 -m pytest -q

Catalog

Blueprint	Category	Purpose
adaptive_experiment_planning	Science	Use this blueprint to choose the next experiment from prior yield, toxicity, cost, and confidence signals before spending lab capacity.
ai_audit_readiness	Security	Use this blueprint to turn AI and cyber control requirements into an evidence-backed readiness pack your team can review before audit or launch.
ai_strategy_planning	Business	Use this blueprint to turn messy discovery notes into a board-ready AI strategy recommendation with priorities, risks, and next steps.
claim_risk_triage	Finance	Use this blueprint to prioritize claim reviews by combining fraud signals, queue pressure, and adjuster capacity into an auditable triage recommendation.
climate_resilience_planning_simulation	Science	Use this blueprint to compare local flood mitigation plans as rainfall, water levels, pump capacity, and vulnerable assets change.
closed_loop_agent_runtime	Engineering	Use this blueprint to understand how a closed-loop agent observes changing work, decides on each tick, and leaves an inspectable action trail.
cluster_reliability_simulation	Engineering	Use this blueprint to test and demo advanced cluster reliability management with Nomad-inspired scheduling, lifecycle, recovery, drain, and service-discovery controls.
codebase_memory_compression_analysis	Engineering	Use this blueprint to see how compressed agent memory helps teams analyze large codebases without flooding every worker with full repository context.
context_memory_audit	Engineering	Use this blueprint to audit how role-specific context packets move through a multi-agent decision process and affect the final recommendation.
context_memory_compression	Engineering	Use this blueprint to test whether compressed working memory keeps a growing LLM workflow useful while reducing context load and cost.
contract_negotiation_simulation	Engineering	Use this blueprint to model buyer-supplier negotiation rounds and compare offers as demand, price, and capacity change.
credit_default_warning	Finance	Use this blueprint to spot borrower default pressure early and compare policy responses before credit risk turns into losses.
customer_lifecycle_email_auto	Business	Use this blueprint to plan, generate, check, send, and track lifecycle emails so customer outreach stays timely, governed, and tied to customer state.
drug_discovery_simulation	Science	Use this blueprint to run a multi-stage drug discovery loop that generates, filters, and evaluates candidates with reviewable evidence.
ecosystem_simulation	Science	Use this blueprint to explore ecosystem interventions across regions and compare population effects before making policy or field decisions.
environment_control_simulation	Engineering	Use this blueprint to experiment with control decisions in a changing environment where load, temperature, reserve, and service levels interact.
event_stream_triage	Engineering	Use this blueprint to triage noisy event streams and see how anomaly pressure, queue depth, and false positives change decisions over time.
facility_safety_video_monitor	Security	Use this blueprint to watch an approved camera stream, detect visible people, summarize what is observable, and raise safety alerts with a reviewable event trail.
human_approval_gate	Security	Use this blueprint to add a human approval checkpoint before an agent applies high-impact actions, with clear review and audit records.
human_review_workflow	Security	Use this blueprint to run a complete human review loop where requests, decisions, revisions, and applied actions are all captured.
liquidity_risk_monitor	Finance	Use this blueprint to monitor market microstructure signals and understand emerging liquidity risk before trades become expensive or fragile.
live_telemetry_monitor	Engineering	Use this blueprint to process live telemetry chunks, detect changing signal patterns, and summarize what operators need to notice.
llm_tool_orchestration	Engineering	Use this blueprint to see an LLM agent call a forecast tool, weigh the result, and choose a resource action you can inspect.
message_routing_trace	Engineering	Use this blueprint to trace how messages move through router and aggregator agents so workflow wiring is easier to debug and explain.
motion_planning_simulation	Science	Use this blueprint to visualize shared-world motion planning so you can inspect coordination, conflicts, and route choices between agents.
native_live_monitor_service	Engineering	Use this blueprint to run a lightweight native monitor that keeps producing decisions over live state until you stop it.
network_threat_monitor	security	This generated blueprint monitors network events, scores suspicious spamware/malware/hack behavior, and writes a dry-run alarm artifact for human review.
openshell_sandbox_worker_pipeline	Engineering	Use this blueprint to run shell and Python workers inside isolated execution boundaries while keeping each step traceable.
outbreak_response_simulation	Science	Use this blueprint to compare outbreak response policies as infections, mobility, vaccination, and hospital load evolve.
parallel_worker_benchmark	Engineering	Use this blueprint to stress-test broad parallel fan-out and measure how deterministic workers behave as runtime scale increases.
policy_feedback_optimization_simulation	Engineering	Use this blueprint to tune policy thresholds through repeated feedback so you can compare rewards, incidents, and tradeoffs before deployment.
portfolio_crash_stress_simulation	Finance	Use this blueprint to stress a portfolio against crash, rate, and liquidity shocks and review rebalancing choices before taking action.
pricing_profit_simulation	Business	Use this blueprint to compare pricing moves against demand, inventory, margin, and competitors before you change prices in the real world.
python_sdk_research_pipeline	Engineering	Use this blueprint to author a MirrorNeuron workflow directly in Python and compile it into a runnable blueprint bundle.
python_sdk_research_service	Engineering	Use this blueprint to run a Python-defined workflow as a long-lived research service with repeated stateful turns.
revenue_retention_simulation	Business	Use this blueprint to compare retention offers for at-risk customers and choose interventions with clear revenue, churn, and customer-experience tradeoffs.
sandboxed_codegen_review	Security	Use this blueprint to generate, review, and validate code inside a sandboxed LLM loop before trusting the result.
service_capacity_simulation	Business	Use this blueprint to test staffing, deflection, and escalation decisions before service queues miss SLA targets.
state_audit_simulation	Engineering	Use this blueprint to record simulation state changes so every agent decision can be inspected after the run.
stream_backpressure_simulation	Engineering	Use this blueprint to observe bounded queues, slow workers, and retry-later behavior before building live stream workflows.
supply_chain_resilience_simulation	Business	Use this blueprint to rehearse supplier disruption responses and choose actions that protect inventory, fulfillment, and customer service levels.
traffic_control_simulation	Science	Use this blueprint to test traffic signal and rerouting controls against speed, incident, volume, and emissions tradeoffs.
user_activity_rmf_triage	general	A local security worker that watches user activity, detects suspicious behavior, asks risky sessions to re-authenticate, and writes RMF/ATO/cATO-ready evidence artifacts.
vendor_selection_decision	Business	Use this blueprint to convert requirements and vendor responses into a scored recommendation, implementation plan, and reviewable decision record.
zip_code_property_memory_ranking	Finance	Use this blueprint to rank property opportunities while preserving useful deal memory across noisy ZIP-code history, broker flow, financing constraints, and past outcomes.
zip_code_property_ranking	Finance	Use this blueprint to rank property acquisition opportunities by ZIP-code demand, price, cap-rate, and risk signals before committing diligence time.

Folder Contract

Most blueprint folders contain:

Path	Purpose
README.md	Self-contained quickstart, inspection notes, and validation guidance.
SPEC.md	User-facing problem, outcome, evaluation criteria, limits, and upgrade path.
TERM.md	Terms, assumptions, or domain notes when present.
manifest.json	Runtime graph, entrypoints, metadata, runners, services, and environment access.
config/default.json	Default launch configuration and mock/sample inputs.
config/overwrite.json	Optional local overrides. Do not commit customer secrets.
payloads/	Worker code, prompts, policies, fixtures, and support files.

Safety Checklist

• Review manifest.json, payloads/, and pass_env before live runs.
• Start with mock, dry-run, or quick-test settings before enabling real external services.
• Keep customer-specific inputs and secrets in local overrides or environment variables.
• Update the local blueprint README and SPEC.md when behavior, inputs, outputs, or limits change.