An AI-powered DeFi intelligence skill providing multi-chain token launch metrics, wallet analytics, and LP strategies paid per request via x402 USDC.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install hyre-defi
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install hyre-defi using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
HYRE Agent is a production-grade decentralized finance intelligence engine engineered to empower AI agents with real-time financial analytics, trading indicators, and cross-chain execution metrics. Operating within the Openclaw Skills ecosystem, the skill eliminates standard API key hurdles through the integration of the x402 HTTP protocol. This allows AI systems to make native microtransactions in USDC, paying strictly per request.
Supporting chains like Solana, Base, and SKALE, the skill opens up 24 high-performance endpoints. These endpoints expose token trenches, early buyer detection, trader profit-and-loss (PnL), liquidity pool recommendation algorithms, and deBridge routing. Developers can integrate HYRE Agent to give their systems deep market awareness and safe execution strategies using advanced model-driven insights.
Ensure you have curl installed on your system. Because this tool runs inside the Openclaw Skills architecture, there is no need for local API keys; all transactions are processed on-demand using standard HTTP requests.
Query the newest token launches:
curl -X POST https://hyreagent.fun/trenches/new-tokens \
-H "Content-Type: application/json" \
-d '{"limit": 5}'
Generate an AI-driven safety verdict and confidence risk score on any token mint:
curl -X POST https://hyreagent.fun/trenches/token-verdict \
-H "Content-Type: application/json" \
-d '{"mint": "TOKEN_MINT_ADDRESS"}'
HYRE Agent features highly standardized data schemas optimized for machine readability and structured parsing.
Route your requests dynamically by prefixing the endpoints:
| Chain | Route Prefix | Example |
|---|---|---|
| Solana | / (root) |
POST /defi/tvl |
| Base | /base/ |
POST /base/defi/tvl |
| SKALE | /skale/ |
POST /skale/defi/tvl |
All endpoints return a uniform payload pattern:
| Field | Type | Description |
|---|---|---|
data |
Object | Raw, endpoint-specific metrics and details |
insight |
String | AI-generated analysis summarizing safety and market trends |
signal |
String | Clear actionable recommendation from the vocabulary |
confidence |
Float | Confidence value ranging from 0.00 to 1.00 |
sources |
Array | Protocol networks checked (e.g., Jupiter, DefiLlama) |
model_used |
String | AI engine resolving the call (e.g., claude-3.5-haiku) |
snipe | watch | avoidfollow | ignoreadd_liquidity | rebalance | holdhigh_yield | low_yieldPOST /lp/pools-strategy to retrieve complete LP allocation guidelines, risk profiling, and yields.POST /ask with natural language to have the backend LLM dynamically dispatch multiple API queries and synthesize the results.POST /trenches/token-snipers during launch trenches.Loading
Discover, verify, and install high-quality AI agent capabilities from skills.sh and ClawHub using the Find Popular Skills utility.

A multi-source RSS and Atom digest generator that aggregates, filters, and summarizes updates from YouTube, GitHub, and Microsoft Tech Community.

An AI-first search and booking integration that finds local events, restaurants, and activities using natural-language queries.

A high-reliability email reputation and deliverability verifier that leverages OpenMerch for agentic workflows.

An agentic stock monitoring skill that provides real-time queries, portfolio analysis, and automated threshold alerts for Chinese (A-shares), Hong Kong, and US stock markets without requiring API keys.

An AI agent cognitive framework that applies Oliver's persona, evidence standards, risk tolerance adjustments, and structured decision models.








































