adr-writer
Generates Architecture Decision Records capturing context, rationale, alternatives, and consequences in numbered status-tracked format. Triggers on: "write an…
Orchestrates business idea validation via parallel sub-agents: Lean Canvas, JTBD, market/competitive/feasibility research, SWOT/PESTLE, and a weighted scorecard with verdict. Triggers on: "validate this idea", "evaluate my startup idea", "is this idea worth pursuing", "score
$ npx -y skills add Mathews-Tom/armory --skill idea-validator --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/idea-validatorContext preview
The summary Claude sees to decide when to auto-load this skill.
Orchestrates business idea validation via parallel sub-agents: Lean Canvas, JTBD, market/competitive/feasibility research, SWOT/PESTLE, and a weighted scorecard with verdict. Triggers on: "validate this idea", "evaluate my startup idea", "is this idea worth pursuing", "score
name: idea-validator description: 'Orchestrates business idea validation via parallel sub-agents: Lean Canvas, JTBD, market/competitive/feasibility research, SWOT/PESTLE, and a weighted scorecard with verdict. Triggers on: "validate this idea", "evaluate my startup idea", "is this idea worth pursuing", "score this business idea", "validate my pitch".' metadata: version: 1.0.1 complements: [market-analyzer, competitive-analyzer, feasibility-assessor] category: review tags: [idea-validation, lean-canvas, jtbd, swot] difficulty: advanced phase: define
Perform rigorous, multi-dimensional validation of business ideas by orchestrating parallel research and synthesizing findings into a scored, actionable report. The goal is honest assessment — surfacing what holds up under scrutiny and what does not — rather than cheerleading or reflexive dismissal.
Classify the input:
| Input Type | Indicators | Adaptation | | ---------------- | ------------------------------------------- | ------------------------------------------- | | Elevator pitch | 1-3 sentences, informal | Extract hypothesis, ask for missing details | | Business plan | Structured, multiple sections | Validate claims against research | | Feature proposal | Technical focus, existing product context | Narrow scope to feature viability | | Repo/codebase | File paths, code references | Technical feasibility + market fit | | Pivot assessment | "Should we pivot to...", comparison context | Compare current vs proposed direction |
Extract core elements from the input:
If any core element is missing or ambiguous, ask clarifying questions. Cap at 5 questions. Do not proceed with fabricated assumptions for missing elements — get real answers.
Build a complete Lean Canvas using `references/lean-canvas.md` as the framework guide.
Populate all 9 boxes:
1. **Problem** — top 3 problems the customer faces 2. **Customer Segments** — target market + early adopter profile 3. **Unique Value Proposition** — single clear compelling message 4. **Solution** — top 3 features addressing the top 3 problems 5. **Channels** — path to reaching customers 6. **Revenue Streams** — pricing model and revenue mechanics 7. **Cost Structure** — fixed and variable costs 8. **Key Metrics** — what to measure for progress 9. **Unfair Advantage** — what cannot be easily copied or bought
For each box:
Apply the JTBD framework from `references/jtbd-framework.md`:
Assess whether the proposed solution addresses the core job better than current alternatives. Identify the four forces of progress: push of current situation, pull of new solution, anxiety of change, habit of status quo.
Use the Agent tool to spawn three research agents in parallel. Each agent receives a focused research brief derived from the idea context established in Phases 1-3. These agents perform the same analysis covered by the complementary skills (`market-analyzer`, `competitive-analyzer`, `feasibility-assessor`) — when those skills are installed, their reference frameworks inform the analysis depth; when running standalone, the agents use the inline briefs below.
Prompt the agent:
> Analyze the market opportunity for [idea summary with target customer and problem context]. Research TAM/SAM/SOM estimates, market growth trends, customer segment sizing, and timing signals. Use WebSearch to find industry reports, market data, analyst projections, and trend indicators. Produce a structured market analysis with quantified estimates and cited sources. Include: market size estimates with methodology, growth rate and trajectory, key market drivers, timing assessment (too early, right time, too late), and customer willingness to pay signals.
Prompt the agent:
> Analyze the competitive landscape for [idea summary with solution and value proposition context]. Discover direct competitors (same solution, same customer), indirect competitors (different solution, same job-to-be-done), and potential competitors (adjacent players who could enter). Compare features, pricing, and positioning across competitors. Assess Porter's Five Forces for this industry. Identify moats and defensibility factors. Use WebSearch for competitor data from Product Hunt, G2, Capterra, Crunchbase, and company sites. Produce a structured competitive analysis with competitor matrix, force scores, and positioning assessment.
Prompt the agent:
> Assess the financial and technical feasibility of [idea summary with solution details and revenue model]. Calculate unit economics using industry benchmarks: customer acquisition cost (CAC), lifetime value (LTV), LTV:CAC ratio, gross margins, payback period. Evaluate technical complexity: architecture requirements, key technical risks, bui
Curated, production-grade skills, agents, hooks, rules, commands, utilities, and presets for AI coding agents. No magic, no demos — battle-tested workflows built for developers who use AI seriously.
Repo: Mathews-Tom/armory
Generates Architecture Decision Records capturing context, rationale, alternatives, and consequences in numbered status-tracked format. Triggers on: "write an…
Build AI agents and automate Claude Code programmatically via the Claude Agent SDK and headless CLI mode. Covers Python SDK, claude -p, SDK MCP servers, hooks,…
Audits and enhances FastAPI and REST API documentation: missing descriptions, response codes, examples, docstrings, Pydantic models, OpenAPI spec. Triggers on:…
Generate architecture diagrams as fully editable SVG with native AWS, Azure, and GCP icons for cloud diagrams, or hand-drawn generic icons for everything else.…
Architecture reviews across 7 dimensions (structural, scalability, enterprise readiness, performance, security, ops, data) with scored reports. Triggers on:…
Optimize and prepare figures for arXiv submission: format conversion (EPS/PDF/PNG/JPG), size reduction, metadata stripping, processor compatibility (DVI vs…