protocol-deep-dive

name	protocol-deep-dive
description	Comprehensive analysis of a specific DeFi protocol — mechanics, risks, tokenomics, and competitive position
tags	["research","protocol","analysis","defi","deep-dive"]
agent	researcher
var	${var} is the protocol name or identifier to analyze. Required. Example: "Jupiter", "Kamino Finance", "Drift Protocol"

protocol-deep-dive

Priority: P2 (on-demand) Data sources: DeFiLlama, CoinGecko, GitHub, protocol docs, governance forums Output: Deep dive report in memory/research/protocols/

Instructions

You are executing the protocol-deep-dive skill for the Researcher agent.

Step 1: Identify Protocol

Resolve ${var} to protocol identifiers:

curl -s "https://api.llama.fi/protocols" | jq '.[] | select(.name | ascii_downcase | contains("{query}"))'

Collect: DeFiLlama slug, CoinGecko ID, GitHub org, official website, docs URL.

Step 2: Core Metrics

TVL and growth:

curl -s "https://api.llama.fi/protocol/{slug}"

Extract: current TVL, TVL history (7d, 30d, 90d trends), chain breakdown, TVL by pool/vault.

Token data (if applicable):

curl -s "https://api.coingecko.com/api/v3/coins/{coingecko_id}?localization=false&tickers=true&market_data=true&community_data=true&developer_data=true"

Extract: price, market cap, FDV, circulating supply, total supply, trading volume.

Step 3: Technical Architecture

From docs and GitHub:

• What blockchain(s) is it deployed on?
• What programming language are the contracts in?
• Is the code audited? By whom? When?
• Is the code open source? License?
• Key dependencies (oracles, bridges, other protocols)

curl -s -H "Authorization: token ${GITHUB_TOKEN}" "https://api.github.com/orgs/{github_org}/repos?sort=pushed&per_page=10"

Step 4: Risk Assessment

Evaluate:

• Smart contract risk: Audit status, code complexity, upgrade mechanisms
• Oracle risk: What price feeds are used? Single point of failure?
• Governance risk: Who controls upgrades? Timelock duration?
• Liquidity risk: How concentrated is TVL? Withdrawal mechanics?
• Regulatory risk: Is the protocol compliant? KYC requirements?

Score each risk dimension 1-5 (1=low risk, 5=critical).

Step 5: Competitive Position

Compare to top 3 competitors in same category:

• TVL comparison
• Feature comparison
• Fee structure comparison
• User count / unique wallets
• Token performance comparison

Step 6: Generate Report

{
  "protocol": "Jupiter",
  "report_date": "2024-01-15",
  "category": "DEX Aggregator",
  "chains": ["Solana"],
  "metrics": {
    "tvl_usd": 800000000,
    "tvl_7d_change_pct": 12.3,
    "daily_volume_usd": 450000000,
    "unique_users_7d": 125000,
    "token_price": 1.23,
    "market_cap": 1500000000,
    "fdv": 12000000000
  },
  "architecture": {
    "contracts_language": "Rust (Anchor)",
    "audited": true,
    "auditors": ["OtterSec", "Neodyme"],
    "open_source": true,
    "oracles": ["Pyth", "Switchboard"]
  },
  "risk_scores": {
    "smart_contract": 2,
    "oracle": 2,
    "governance": 3,
    "liquidity": 1,
    "regulatory": 3,
    "overall": 2.2
  },
  "competitive_position": "Market leader in Solana DEX aggregation",
  "strengths": ["Best routing", "Highest volume", "Active development"],
  "weaknesses": ["Token FDV concern", "Concentration risk"],
  "opportunities": ["Perps growth", "Cross-chain expansion"],
  "threats": ["New aggregators", "Direct DEX competition"]
}

Write to memory/research/protocols/{protocol_slug}.json and post to analyst mesh.

Anti-Patterns

• Do NOT copy marketing material as analysis. Be critical and objective.
• Do NOT skip risk assessment. Every protocol has risks.
• Do NOT present opinions as facts. Quantify where possible.

Exit Codes

• SKILL_OK — deep dive complete
• SKILL_PARTIAL — some data sources unavailable
• SKILL_FAIL — could not identify or analyze protocol

Output

Commit message format: researcher: protocol-deep-dive — {protocol_name} (risk: {overall_score}/5)