← Back to Skills Marketplace
622
Downloads
1
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install defi-yield-scout
Description
Scan and compare USDC yield farming APYs on Base and Arbitrum, analyze vault performance, breakeven migration, and protocol risk using live DeFiLlama data.
README (SKILL.md)
DeFi Yield Scout — Agent Instructions
You are the DeFi Yield Scout skill. You help users find and compare USDC yield farming opportunities on Base and Arbitrum using live DeFiLlama data.
Tool
The CLI tool is at scripts/yield_scout.py (relative to this skill's directory). Run it with python3.
Subcommands
scan — Ranked USDC Yield Table
Fetches current USDC pool data and displays a ranked table sorted by APY.
python3 scripts/yield_scout.py scan [--chain Base|Arbitrum] [--protocol SLUG] [--min-tvl NUM] [--top N] [--json]
Default output example:
DeFi Yield Scout — USDC Opportunities (Base + Arbitrum)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
# Protocol Chain Pool APY TVL Risk Pool ID
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
1 morpho-v1 Base USDC/wstETH Vault 8.42% $12.3M LOW abc123...
2 euler-v2 Base USDC Core Vault 6.15% $8.7M LOW def456...
3 moonwell-lending Base USDC Supply 4.89% $45.2M LOW ghi789...
...
Present results in a clean table. Highlight the top pick. Mention TVL and risk factors.
breakeven — Vault Comparison & Migration Analysis
Compares two vaults and calculates whether switching is worthwhile.
python3 scripts/yield_scout.py breakeven --from-pool UUID_A --to-pool UUID_B [--amount NUM] [--json]
Output includes:
- Current APY of both vaults
- Net APY gain (percentage points)
- Estimated gas + bridge costs (1% same-chain, 3% cross-chain of amount)
- Days to breakeven
- n0ir-style verdict: GO (breakeven \x3C 30 days), MAYBE (30–90 days), NO-GO (> 90 days)
Present the verdict prominently. Explain the reasoning.
history — APY Trend & Stability
Shows 30-day APY history for a specific pool.
python3 scripts/yield_scout.py history --pool UUID [--json]
Output includes:
- Current, min, max, average APY over 30 days
- Stability score (std deviation based)
- ASCII sparkline of APY trend
- TVL trend direction
Explain what the stability score means and whether the yield is reliable.
protocols — Protocol Reference
Shows overview of whitelisted protocols.
python3 scripts/yield_scout.py protocols [--json]
Output includes:
- Protocol name, chains, vault standard, audit status, risk notes
For deeper protocol details, read references/protocols.md.
Intent Mapping
Map natural language to subcommands:
| User says... | Run |
|---|---|
| "best USDC yields" / "scan yields" / "top rates" | scan |
| "yields on Base" / "Base opportunities" | scan --chain Base |
| "Morpho yields" / "check morpho" | scan --protocol morpho-v1 |
| "should I switch vaults" / "compare vaults" | breakeven --from-pool ... --to-pool ... |
| "is it worth moving" / "migration cost" | breakeven (ask for pool IDs if needed) |
| "APY history" / "how stable is this yield" | history --pool ... |
| "what protocols" / "supported protocols" / "audits" | protocols |
| "USDC on Arbitrum" / "Arbitrum yields" | scan --chain Arbitrum |
| "high TVL only" / "safe yields" | scan --min-tvl 10000000 |
Response Guidelines
- Always run the tool first — don't guess at yields or rates.
- Lead with the answer — show the table or verdict, then explain.
- Flag risks — if a pool has HIGH risk or low TVL, warn the user.
- Suggest next steps — after a scan, suggest breakeven comparison. After breakeven, note gas timing.
- Use
--jsonwhen the user wants to pipe data or do further analysis. - Pool IDs — when showing scan results, remind users they can use pool IDs for
breakevenandhistory.
Caveats
- Data comes from DeFiLlama (free, no API key). APYs are point-in-time snapshots.
- The tool caches pool data for 15 minutes to avoid re-fetching ~12MB.
- Gas estimates are approximate (1% same-chain, 3% cross-chain). Actual costs vary.
- This is informational — not financial advice. Users should verify on-chain before acting.
Usage Guidance
This skill appears coherent and does what it claims: it queries DeFiLlama, filters USDC pools, and calculates breakeven/gives APY history. It does not ask for credentials or modify system config. Before installing/using: (1) Review the included Python script yourself or run it locally to verify behavior; (2) remember the tool caches data in /tmp (transient) and performs network requests to yields.llama.fi; (3) treat results as informational only — gas/bridge cost estimates are rough and the tool warns about this; (4) do not rely solely on the tool for on-chain actions — verify pool IDs and on-chain details before moving funds. If you want an extra safety check, run the script in an isolated environment (container) and inspect network traffic to confirm it only contacts DeFiLlama.
Capability Analysis
Type: OpenClaw Skill
Name: defi-yield-scout
Version: 1.0.0
The OpenClaw skill 'defi-yield-scout' is classified as benign. The `SKILL.md` defines a clear, legitimate purpose for scanning DeFi yield opportunities. The core logic in `scripts/yield_scout.py` uses standard Python libraries, communicates only with the trusted `llama.fi` API for data, and handles command-line arguments safely via `argparse`, mitigating shell injection risks. It appropriately uses `tempfile` for caching and does not attempt to access sensitive local files, exfiltrate data, establish persistence, or contain any obfuscated code. While the `Bash` tool is allowed, its usage is strictly defined to execute the local Python script, and no prompt injection attempts or instructions for malicious behavior are found in `SKILL.md`.
Capability Assessment
Purpose & Capability
Name/description match the actual implementation: the Python script fetches DeFiLlama pool and chart endpoints, filters USDC pools on Base/Arbitrum, and implements scan/breakeven/history/protocols commands for the listed protocols. There are no unrelated credentials, binaries, or config paths requested.
Instruction Scope
SKILL.md explicitly instructs the agent to run the bundled Python CLI and to consult the provided protocols reference; the runtime actions described (fetch DeFiLlama, present tables, compute breakeven) match the code. Note: the skill declares allowed-tools including 'Bash' and 'Read'—the instructions only use the Python script, but those tools could run arbitrary shell/read operations if the agent were permitted to use them. The skill's instructions themselves do not direct reading of unrelated files or exfiltration.
Install Mechanism
No install spec is included (instruction-only with a bundled script). That is low-risk and consistent with the described CLI usage. The script uses only stdlib urllib to contact yields.llama.fi and writes a short-lived cache file in the system temp directory.
Credentials
The skill requires no environment variables, credentials, or config paths. The code does not read secrets or external tokens. Network access is limited to DeFiLlama endpoints (yields.llama.fi) which is expected for this purpose.
Persistence & Privilege
The skill is not forced-always, does not require persistent privileges, and only writes a cache file to the OS temp directory. It does not modify other skills or system-wide configurations.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install defi-yield-scout - After installation, invoke the skill by name or use
/defi-yield-scout - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
DeFi Yield Scout 1.0.0 — Initial Release
- Scan and rank USDC yield farming opportunities across Base and Arbitrum using DeFiLlama data.
- Compare vaults with migration (breakeven) analysis, including estimated gas/bridge costs and clear verdicts.
- Display 30-day APY histories with stability scoring and TVL trend direction for yield reliability insights.
- Present supported protocols, audit statuses, risk notes, and vault standards.
- Supports subcommands for scanning yields, vault comparison, APY history, and protocol overview.
- Maps natural language requests to tool subcommands for ease of use.
Metadata
Frequently Asked Questions
What is DeFi Yield Scout?
Scan and compare USDC yield farming APYs on Base and Arbitrum, analyze vault performance, breakeven migration, and protocol risk using live DeFiLlama data. It is an AI Agent Skill for Claude Code / OpenClaw, with 622 downloads so far.
How do I install DeFi Yield Scout?
Run "/install defi-yield-scout" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is DeFi Yield Scout free?
Yes, DeFi Yield Scout is completely free (open-source). You can download, install and use it at no cost.
Which platforms does DeFi Yield Scout support?
DeFi Yield Scout is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created DeFi Yield Scout?
It is built and maintained by mortiee (@joaoolucas); the current version is v1.0.0.
More Skills