docs: enhance features documentation with multi-agent orchestration details and add new use case for incident triage

Author: nmaguiar

features.md

@@ -4,7 +4,7 @@ title: Features
 permalink: /features/
 ---
 
-mini-a packs a comprehensive set of features into a minimalist framework. This page covers everything from model selection and cost optimization to security, tooling, and output formats.
+mini-a packs a comprehensive set of features into a minimalist framework. This page covers everything from model selection and cost optimization to multi-agent orchestration, security, tooling, and output formats.
 
 ---
 
@@ -215,10 +215,12 @@ Other agents or applications can then delegate tasks to this worker instance.
 
 ---
 
-## Planning & Delegation
+## Planning & Multi-Agent Delegation
 
 For complex goals, mini-a can **plan** a series of steps before executing them, and optionally **delegate** subtasks to child agents or remote workers.
 
+Think of delegation as an orchestration layer: one parent agent manages task decomposition, parallel execution, and result aggregation across multiple workers.
+
 ### Planning Styles
 
 | Style | Description |
@@ -243,6 +245,21 @@ mini-a usedelegation=true
 
 Delegation also works with remote workers — the main agent can send subtasks to mini-a instances running in worker mode on other machines.
 
+### Orchestration Pattern
+
+Use this pattern when you want mini-a to coordinate multiple agents for larger workloads:
+
+1. Start one or more worker instances (`mini-a worker=true workerport=9090`).
+2. Run a parent agent with planning and delegation enabled.
+3. Set concurrency limits so execution stays predictable.
+4. Let the parent combine worker outputs into a single final result.
+
+```bash
+mini-a useplanning=true planstyle=validate usedelegation=true \
+  workers='http://worker1:9090,http://worker2:9090' maxconcurrent=4 \
+  goal='Analyze this monorepo, group findings by domain, and produce one prioritized action plan'
+```
+
 ---
 
 ## Chatbot Mode

index.md

@@ -8,8 +8,8 @@ permalink: /
   <p class="tagline">Your goals. Your LLM. One command.</p>
   <p class="hero-description">
     A minimalist autonomous agent framework built on OpenAF. Connect any LLM,
-    use 20+ built-in MCP tool servers, and let mini-a figure out how to
-    achieve your goals — from the terminal, a web UI, or your own code.
+    use 20+ built-in MCP tool servers, and orchestrate multi-agent delegation
+    for complex goals — from the terminal, a web UI, or your own code.
   </p>
   <div class="cta-buttons">
     <a href="{{ '/getting-started' | relative_url }}" class="btn btn-primary">Get Started</a>
@@ -72,6 +72,10 @@ permalink: /
     <h3>20+ Tools, Zero Code</h3>
     <p>Built-in MCP servers for time, finance, databases, web, email, Kubernetes, and more — ready to use.</p>
   </div>
+  <div class="feature-card">
+    <h3>Multi-Agent Orchestration</h3>
+    <p>Enable delegation to split goals into subtasks and run them in parallel across child agents or remote workers.</p>
+  </div>
   <div class="feature-card">
     <h3>40-60% Fewer Tokens</h3>
     <p>Automatic context optimization, conversation compaction, and smart summarization keep costs low.</p>

use-cases.md

@@ -6,7 +6,7 @@ permalink: /use-cases/
 
 # Real-World Use Cases
 
-mini-a is designed to be flexible enough for a wide range of tasks. Below are eight practical scenarios that demonstrate how to leverage the framework in real-world workflows, from DevOps automation to cost-optimized enterprise deployments.
+mini-a is designed to be flexible enough for a wide range of tasks. Below are nine practical scenarios that demonstrate how to leverage the framework in real-world workflows, from DevOps automation to cost-optimized enterprise deployments.
 
 ---
 
@@ -259,7 +259,40 @@ When triggered as part of a CI/CD pipeline (e.g., on pull request creation), the
 
 <div class="use-case">
 
-## 8. Cost Optimization: Enterprise Dual-Model Strategy
+## 8. Multi-Agent Orchestration: Distributed Incident Triage
+
+### Context
+
+When incident volume spikes, triage work often spans multiple systems: logs, tickets, deployment history, and service dashboards. mini-a delegation lets a parent agent orchestrate specialized child agents in parallel, then merge their outputs into one coordinated response plan.
+
+### Command
+
+```bash
+mini-a useplanning=true planstyle=validate usedelegation=true \
+  workers='http://worker1:9090,http://worker2:9090' maxconcurrent=4 \
+  goal='Investigate the last 24h production incidents, identify top root causes, and propose a prioritized remediation plan'
+```
+
+### How It Works
+
+1. The parent agent builds a validated plan (`planstyle=validate`) and identifies independent workstreams.
+2. Delegation dispatches sub-tasks to local child agents and remote workers listed in `workers`.
+3. Worker agents run in parallel (bounded by `maxconcurrent`) and return structured intermediate results.
+4. The parent agent reconciles overlaps, resolves conflicts, and produces one final triage report with next actions.
+
+### Expected Behavior
+
+The final output contains grouped root causes (for example deploy regressions, infra saturation, and dependency outages), confidence levels, and a prioritized remediation queue with owners and suggested follow-up checks. Compared to fully serial triage, parallel delegation shortens time-to-report while keeping one coherent, centralized result.
+
+<div class="screenshot-placeholder">[SCREENSHOT-PLACEHOLDER: S23 -- Delegation orchestration dashboard/report]</div>
+
+</div>
+
+---
+
+<div class="use-case">
+
+## 9. Cost Optimization: Enterprise Dual-Model Strategy
 
 ### Context
 
@@ -320,7 +353,7 @@ Estimated Cost Savings: ~45% vs single-model deployment
 
 This visibility helps teams fine-tune their model selection and monitor spend over time.
 
-<div class="screenshot-placeholder">[SCREENSHOT-PLACEHOLDER: S23 -- Metrics showing cost savings]</div>
+<div class="screenshot-placeholder">[SCREENSHOT-PLACEHOLDER: S24 -- Metrics showing cost savings]</div>
 
 </div>