defillama/defillama-skills/risk-assessment

Claude Code and agent skill by defillama/defillama-skills

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Protocol and DeFi risk evaluation covering hack history, oracle dependencies, treasury health, TVL concentration, and yield sustainability. Use when the user asks "is X safe", "how risky is", protocol security, risk analysis, or wants to evaluate risk before investing or depositing funds.

Install

npx vskill@latest install defillama/defillama-skills/risk-assessment

SKILL.md

Risk Assessment

Evaluate the risk profile of a DeFi protocol by examining its security history, oracle dependencies, treasury, fundamentals, and yield levels.

Workflow

Step 1 - Resolve the protocol entity
defillama:resolve_entity
entity_type: "protocol"
name: "<user-provided name>"
Step 2 - Hack history

Check whether the protocol has been exploited before.

defillama:get_events
protocol: "<slug>"
event_type: "hacks"

Any past hacks are a significant risk signal. Note the date, amount lost, and whether funds were recovered.

Step 3 - Oracle dependencies

Identify which oracle the protocol relies on and how much value it secures.

defillama:get_oracle_metrics

Filter results for the protocol's oracle. Oracle metrics automatically filter to tvl_component = 'base' and aggregate with SUM/GROUP BY to exclude double-counted TVL. A protocol using a small or unproven oracle has higher risk than one using Chainlink or Pyth.

Step 4 - Treasury health

Assess the protocol's financial reserves.

defillama:get_treasury
treasury: "<slug>"

A healthy treasury provides a safety net for bug bounties, insurance, and continued development.

Step 5 - Protocol fundamentals

Check TVL, revenue, and trends to evaluate sustainability.

defillama:get_protocol_metrics
protocol: "<slug>"

Key signals: Is TVL growing or declining? Is the protocol generating real revenue?

Step 6 - Yield analysis

Examine pool APYs for sustainability red flags.

defillama:get_yield_pools
protocol: "<slug>"

Risk Signals

Evaluate each factor and assign a risk level:

SignalRisk LevelExplanation
Recent hack (< 1 year)HIGHProtocol was recently exploited
Past hack (> 1 year), no recurrenceMODERATEWas exploited but has since hardened
No hack historyLOWNo known exploits
No oracle or small oracle TVSMODERATEOracle risk, potential manipulation
Uses Chainlink/Pyth with high TVSLOWBattle-tested oracle infrastructure
Treasury < $1MHIGHNo meaningful safety net
Treasury > $10MLOWStrong financial reserves
APY > 100% from rewards onlyHIGHLikely unsustainable, token emissions
APY from real yield (fees)LOWSustainable yield source
TVL declining > 20% in 30dMODERATEUsers are leaving
Revenue declining with TVLHIGHFundamentals deteriorating

Output Format

Present the report with these sections:

  1. Risk Summary - Overall risk rating (LOW / MODERATE / HIGH) with

a one-sentence justification.

  1. Security History - Hack incidents, amounts, and outcomes.
  2. Oracle Risk - Which oracle is used, TVS secured, reliability.
  3. Treasury Health - Total treasury value, composition, runway.
  4. Fundamental Health - TVL trend, revenue, user activity direction.
  5. Yield Sustainability - Are yields from real fees or emissions?
  6. Risk Factors - Bullet list of all identified risk signals with

their severity level.

Tips

  • No data from get_events for hacks is a positive signal, not missing data.
  • A protocol with high TVL but zero revenue may be subsidizing usage

with token emissions - flag this.

  • Compare treasury value to TVL: treasury < 1% of TVL means limited

ability to cover losses.

  • Multiple past hacks on the same protocol is a stronger negative signal

than a single incident.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/defillama/defillama-skills/risk-assessment/badge)](https://verified-skill.com/skills/defillama/defillama-skills/risk-assessment)

Something wrong with this skill or its page? Tell us.