Skip to content

diegosouzapw/awesome-omni-skill / skills/ai-agents/agent-extend

agent-extend

Verified

Extend the Strategy Coach agent with new capabilities. Use when user says "add coaching phase", "new research pillar", "extend agent", "add strategic output", or asks to enhance the AI coaching methodology.

Strategy Coach Agent Extension Skill

Extends the Strategy Coach agent following the Product Strategy Research Playbook methodology.

When to Use

Activate when user requests:

  • "add coaching phase"
  • "new research pillar"
  • "extend agent"
  • "add strategic output"
  • Enhancing AI coaching methodology

Agent Location

All agent files are in src/lib/agents/strategy-coach/:

FilePurpose
index.tsMain agent interface, Claude API integration
framework-state.tsState tracking, progress calculations
system-prompt.tsMethodology documentation, coaching tone
client-context.tsOrganization context loading

Current Methodology

Four Coaching Phases

  1. Discovery - Understanding context and goals
  2. Research - Exploring the three pillars
  3. Synthesis - Developing strategic hypotheses
  4. Planning - Creating actionable outputs

Three Research Pillars

  1. Macro Market - Industry trends, competition, technology
  2. Customer - Segments, JTBD, pain points
  3. Colleague - Internal capabilities, constraints

Strategic Bets Format

We believe [trend/customer need]
Which means [opportunity/problem space]
So we will explore [hypothesis/initiative direction]
Success looks like [leading indicator metric]

Extension Patterns

Adding a New Phase

  1. Update framework-state.ts:
export interface FrameworkState {
  phase: 'discovery' | 'research' | 'synthesis' | 'planning' | 'new_phase';
  // ...
}

export function calculateProgress(state: FrameworkState): number {
  // Add new phase to progress calculation
}
  1. Update system-prompt.ts:
const NEW_PHASE_GUIDANCE = `
## New Phase: {Phase Name}

### Purpose
{What this phase accomplishes}

### Key Activities
- {Activity 1}
- {Activity 2}

### Transition Criteria
Move to next phase when:
- {Criterion 1}
- {Criterion 2}
`;

Adding a Research Pillar

  1. Update framework-state.ts:
export interface ResearchProgress {
  macroMarket: PillarProgress;
  customer: PillarProgress;
  colleague: PillarProgress;
  newPillar: PillarProgress;  // Add new pillar
}

interface PillarProgress {
  started: boolean;
  completed: boolean;
  insights: string[];
}
  1. Update system-prompt.ts:
const NEW_PILLAR_SECTION = `
### Pillar 4: {Pillar Name}

#### Core Questions
- {Question 1}
- {Question 2}

#### Key Areas to Explore
| Area | Focus |
|------|-------|
| {Area 1} | {Focus description} |
| {Area 2} | {Focus description} |

#### Insight Format
Capture insights as:
- **Observation**: What was discovered
- **Implication**: What it means for strategy
- **Evidence**: Supporting data points
`;

Adding Strategic Output Types

  1. Update framework-state.ts:
export interface CanvasCompletion {
  marketReality: boolean;
  customerInsights: boolean;
  organizationalContext: boolean;
  strategicSynthesis: boolean;
  strategicContext: boolean;
  newOutputType: boolean;  // Add new output
}
  1. Update system-prompt.ts:
const NEW_OUTPUT_TEMPLATE = `
## {Output Name} Template

### Purpose
{What this output provides}

### Structure
1. **Section 1**: {Description}
2. **Section 2**: {Description}

### Example
{Concrete example of the output}
`;

Coaching Tone Guidelines

When extending the agent, maintain:

  • Socratic questioning - Guide through questions, don't prescribe
  • Empathetic acknowledgment - Validate challenges
  • Structured exploration - Follow methodology phases
  • Actionable synthesis - End with clear next steps

Example prompt patterns:

const QUESTIONING_PATTERNS = [
  "What patterns are you seeing in...",
  "How might this affect...",
  "What would need to be true for...",
  "Who else in your organization might have insight on...",
];

State Persistence

State is stored in framework_state JSONB column in conversations table.

Update state through the agent:

export function updateFrameworkState(
  currentState: FrameworkState,
  update: Partial<FrameworkState>
): FrameworkState {
  return {
    ...currentState,
    ...update,
    lastUpdated: new Date().toISOString(),
  };
}

Testing Extensions

Add tests in tests/unit/lib/agents/strategy-coach/:

describe('New Phase', () => {
  it('should calculate progress correctly', () => {
    const state = createStateWithNewPhase();
    expect(calculateProgress(state)).toBe(expectedProgress);
  });

  it('should transition from previous phase', () => {
    // Test phase transition logic
  });
});
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 468kUpdated Claude CodeCodexUnverifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 319kUpdated Claude CodeCodexVerifiedInstall