PIL MCP Server

Pet Intelligence for AI Agents

Connect Claude Desktop or Claude Code to PIL. 6 tools providing structured, weekly-updated, source-grounded pet-commerce intelligence. Currently in beta. Tested with Claude Desktop and Claude Code.

Installation

Claude Desktop

Add to your claude_desktop_config.json:

{ "mcpServers": { "pil": { "command": "npx", "args": ["@pil/mcp-server"], "env": { "PIL_API_KEY": "your-api-key", "PIL_BASE_URL": "https://pil.brandthrive.ai" } } } }

Claude Code

Add to your project or user MCP settings:

# In your project's .mcp.json or ~/.claude/mcp.json { "mcpServers": { "pil": { "command": "npx", "args": ["@pil/mcp-server"], "env": { "PIL_API_KEY": "your-api-key", "PIL_BASE_URL": "https://pil.brandthrive.ai" } } } }

Local Development

# Clone and build git clone [repo] cd packages/pil-mcp-server npm install npm run build # Run directly PIL_API_KEY=your-key PIL_BASE_URL=https://pil.brandthrive.ai node dist/index.js

Available Tools

6 intelligence tools

pil_opportunity_scan10 credits

Full PPOS opportunity assessment (0-100 score). Combines demand, gaps, pricing, readiness, and trends.

When to use: User asks about pet product viability, market entry, or which category to pursue.

Example call and output
// Input { "category": "outdoor_adventure", "market": "UK" } // Output (abbreviated) PPOS 66/100 (Promising). Demand 100, Gap 93, Price 0. Strong demand, clear gaps, needs pricing validation.
pil_demand_assessment5 credits

Buyer demand confidence, signal volume, purchaser weighting, emotion drivers, and trend direction.

When to use: User asks about buyer interest, demand signals, or emotional purchase drivers for pet products.

Example call and output
// Input { "category": "play_enrichment", "market": "NA" } // Output (abbreviated) Demand Confidence 72/100 (Medium-High). 344 signals, 35% purchaser-weighted. Primary emotion: pride. Trend: growing +22%.
pil_competitor_gap_report5 credits

Competitive whitespace where demand exists but products do not address it.

When to use: User asks about competitive gaps, unmet needs, or positioning opportunities.

Example call and output
// Input { "category": "style_identity", "market": "UK" } // Output (abbreviated) 3 gaps found. Top: Comfort gap (confidence 85%), 12 demand signals, 0/20 bestsellers address it.
pil_buyer_language3 credits

Consumer voice themes, top phrases, say-this/don't-say-this messaging guidance.

When to use: User needs product copy, ad messaging, or wants to understand how pet buyers talk.

Example call and output
// Input { "category": "emotional_memorial", "market": "NA" } // Output (abbreviated) 280 signals, 42% purchaser. Top theme: quality_complaints (pain_point). Say: 'love it', 'best purchase'. Don't say: 'military-grade'.
pil_price_benchmark3 credits

Price range, clusters (entry/mid/premium), top bestsellers, cross-market comparison.

When to use: User asks about pricing strategy, competitive prices, or what pet products cost.

Example call and output
// Input { "category": "comfort_living", "market": "FR" } // Output (abbreviated) Median EUR 24.99. Entry: 8-16 (7 products). Mid: 17-32. Premium: 33-48. Top seller: OrthoPet Bed (EUR 29.99, 4.5 stars).
pil_market_alerts3 credits

Week-over-week changes: new bestsellers, price shifts, demand spikes, competitor movements.

When to use: User asks what changed in the pet market, about new competitors, or recent trends.

Example call and output
// Input { "category": "pet_tech", "market": "UK" } // Output (abbreviated) 4 alerts. [HIGH] New entry: SmartFeeder Pro at #3. [MED] Price drop -12% on PetCam X2.

Supported Inputs

Categories

food_nutrition
health_wellness
play_enrichment
style_identity
hygiene_cleaning
emotional_memorial
outdoor_adventure
comfort_living
pet_tech

Markets

UK United Kingdom
FR France / Europe
NA North America

Why PIL over web search

Structured intelligence:Pre-computed weekly signals, not raw search results an agent must parse.
Proprietary framework:PPOS scoring, purchaser weighting, silence detection. Not generic analysis.
7 sources in 1 call:Reddit, Amazon reviews, YouTube, trade publications, Zooplus, Chewy, Etsy. Not 7 separate searches.
Source-grounded:Confidence scores, data quality flags, and explicit limitations. Not ungrounded claims.
Decision-ready:Signal, Meaning, Action in every response. Not information that requires further interpretation.