Vault Intelligence

Vault discovery, comparison, and analytics for Paradex vaults. Ranks vaults by risk-adjusted returns, analyzes operator track records, monitors TVL changes,...

MIT-0 · Free to use, modify, and redistribute. No attribution required.
0 · 5 · 0 current installs · 0 all-time installs
bySergey Vidyuk@sv
MIT-0
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The name/description (Paradex vault discovery, ranking, monitoring) matches the SKILL.md behavior: all runtime instructions describe reading Paradex MCP vault data (paradex_vaults, paradex_vault_summary, etc.) and producing analyses and recommendations. No unrelated binaries, credentials, or external services are requested.
Instruction Scope
SKILL.md restricts runtime actions to calling the listed Paradex MCP tools, running JMESPath filters, computing scores, and presenting results. It does not instruct the agent to read local files, environment variables, system state, or post data to unexpected endpoints. It asks the agent to gather user preferences (risk tolerance, horizon) which is appropriate for recommendation output.
Install Mechanism
No install spec or code files beyond documentation — this is instruction-only. Nothing will be downloaded or written to disk as part of an install step.
Credentials
The skill declares no environment variables, credentials, or config paths and the instructions do not reference any secrets or external API keys. The requested access is proportionate to its stated purpose.
Persistence & Privilege
always:false (default) and autonomous invocation allowed (platform default). The skill does not request permanent presence, nor does it attempt to modify other skills or system-wide settings. Note: autonomous invocation is normal; combine with other red flags only if present (none here).
Assessment
This skill is internally coherent: it’s an instruction-only analytics tool that uses Paradex MCP data tools and asks for no secrets or installs. Before installing, consider: 1) source/maintainer unknown — if you require traceability, prefer skills with a known repo or homepage; 2) confirm your platform provides the listed Paradex MCP tools (paradex_vaults, paradex_vault_summary, etc.), otherwise the skill will fail; 3) it provides deposit recommendations (financial advice) but does not request wallet keys or execute transactions — do not assume it will perform deposits automatically; 4) review outputs critically (scoring methods and risk math are documented but simplified). If you want to proceed, test on read-only queries first and verify the tool responses and data freshness.

Like a lobster shell, security has layers — review code before you run it.

Current versionv1.0.0
Download zip
latestvk974pxhs5byamz0g5w4969hc7s840a4c

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

SKILL.md

Paradex Vault Intelligence

Turns raw vault data from Paradex MCP into investment-grade vault analytics. Helps users discover, compare, and monitor vaults for passive income strategies.

Available MCP Tools (data sources)

ToolWhat it gives youKey params
paradex_vaultsVault config, owner, status, kindvault_address, jmespath_filter
paradex_vault_summaryPerformance metrics (ROI, PnL, drawdown, volume, TVL)vault_address, jmespath_filter
paradex_vault_positionsCurrent open positionsvault_address
paradex_vault_balanceAvailable/locked/total balancevault_address
paradex_vault_account_summaryAccount health, margin, exposurevault_address
paradex_vault_transfersDeposit/withdrawal historyvault_address

Capabilities

1. Vault Discovery & Screening

Use paradex_vault_summary with JMESPath to screen the full vault universe:

By performance:

# Top 5 by total ROI
"sort_by([*], &to_number(total_roi))[-5:]"

# Profitable in last 7 days
"[?to_number(roi_7d) > `0`]"

# Best 30-day performers
"sort_by([*], &to_number(roi_30d))[-10:]"

By risk:

# Lowest max drawdown (safer vaults)
"sort_by([*], &to_number(max_drawdown))[:5]"

# Low recent drawdown + positive returns
"[?to_number(max_drawdown_30d) < `0.05` && to_number(roi_30d) > `0`]"

By size/activity:

# Largest by TVL
"reverse(sort_by([*], &to_number(tvl)))"

# Most active by 24h volume
"reverse(sort_by([*], &to_number(volume_24h)))"

# Most depositors (social proof)
"reverse(sort_by([*], &num_depositors))"

2. Vault Deep Dive

For a specific vault, gather comprehensive data:

  1. paradex_vaults — get config: owner, kind, status, creation details
  2. paradex_vault_summary — performance: ROI (24h/7d/30d/total), PnL, drawdowns, volume, TVL, token price
  3. paradex_vault_positions — current holdings: which markets, sizes, unrealized PnL
  4. paradex_vault_account_summary — account health: margin usage, leverage, exposure
  5. paradex_vault_balance — cash position: available vs. locked
  6. paradex_vault_transfers — fund flows: deposit/withdrawal patterns

3. Vault Comparison

When comparing 2+ vaults, build a comparison matrix:

MetricVault AVault BBetter
Total ROIX%Y%✓ higher
30d ROIX%Y%✓ higher
Max DrawdownX%Y%✓ lower
Sharpe-like ratioROI/DDROI/DD✓ higher
TVL$X$Ycontext-dependent
# DepositorsNM✓ higher (social proof)
24h Volume$X$Y✓ higher (activity)
Position countNMcontext-dependent

Risk-adjusted ranking: Compute a simple Sharpe-like ratio: total_roi / max_drawdown (higher = better risk-adjusted returns). For vaults with zero drawdown, use total_roi alone but flag as "insufficient drawdown history."

4. Vault Risk Assessment

For each vault, assess and report:

  • Drawdown risk: max_drawdown vs. max_drawdown_30d — is drawdown getting worse?
  • Concentration risk: from positions — how many markets, what % of exposure is in largest position?
  • Leverage risk: from account_summary — current leverage vs. available margin
  • Liquidity risk: from TVL + transfers — is TVL stable, growing, or declining?
  • Operator activity: from volume — is the vault actively trading or dormant?

Risk score (1-5):

  • 1 (Low): Low drawdown, diversified positions, moderate leverage, stable/growing TVL
  • 3 (Medium): Some drawdown history, concentrated in 2-3 markets, moderate leverage
  • 5 (High): Large drawdowns, single-market concentration, high leverage, declining TVL

5. Vault Monitoring & Alerts

When asked to monitor a vault, check for:

  • ROI dropping below a threshold
  • Drawdown exceeding user-defined limit
  • TVL declining (depositor exodus)
  • Position concentration changing significantly
  • New large positions opened (strategy shift)

6. Vault Recommendation Engine

When a user asks "which vault should I deposit in?", gather their preferences:

Risk tolerance:

  • Conservative: prioritize low drawdown, stable returns, high TVL, many depositors
  • Moderate: balance ROI and drawdown, accept some concentration
  • Aggressive: prioritize highest ROI, accept higher drawdown and concentration

Time horizon:

  • Short-term: weight roi_24h and roi_7d more heavily
  • Medium-term: weight roi_30d and last_month_return
  • Long-term: weight total_roi and max_drawdown

Then screen, score, and present top 3-5 vaults with reasoning.

Output Format

Vault Screening Results

## Paradex Vault Screening — [criteria]

Found N vaults matching criteria. Top picks:

### 1. [Vault Name/Address]
- ROI: 24h X% | 7d X% | 30d X% | Total X%
- Risk: Max DD X% | 30d DD X%
- Size: TVL $X | Depositors: N
- Activity: 24h Vol $X
- Risk Score: X/5

[1-sentence take on this vault]

Vault Deep Dive

## Vault Analysis — [address]

### Performance
[ROI table across timeframes]

### Current Positions
[Position breakdown with market, size, unrealized PnL]

### Account Health
[Margin usage, leverage, available capacity]

### Risk Assessment
[Risk score with reasoning]

### Fund Flows
[Recent deposit/withdrawal trends]

Important Caveats

  • Past vault performance does not guarantee future results — state this clearly
  • Vault token price can decline — depositors can lose money
  • Withdrawal lockup periods apply — mention the vault's specific lockup
  • This is analysis and screening, not investment advice
  • Vault operator strategies are private — positions give clues but not full picture
  • Small TVL vaults may have inflated ROI percentages from small base effects

See scoring.md for detailed risk scoring methodology and JMESPath query cookbook.

Files

2 total
Select a file
Select a file to preview.

Comments

Loading comments…