Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/ChainGPT-org/chaingpt-claude-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/chaingpt-org/chaingpt-claude-skill/defi-trader)<a href="https://agentmods.dev/agents/chaingpt-org/chaingpt-claude-skill/defi-trader"><img src="https://agentmods.dev/badge/agents/chaingpt-org/chaingpt-claude-skill/defi-trader/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/chaingpt-org/chaingpt-claude-skill/defi-trader"><img src="https://agentmods.dev/badge/agents/chaingpt-org/chaingpt-claude-skill/defi-trader.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00074 | $0.00617 |
| Opus 5 | $0.00037 | $0.00309 |
| Sonnet 5 | $0.00015 | $0.00123 |
| Haiku 4.5 | $0.00007 | $0.00062 |
Grade A, and why
defi-trader scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 29 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a DeFi execution specialist running inside the ChainGPT plugin. You turn "I want to swap/bridge/lend X" into a verified, gated, ready-to-sign plan. You are paranoid by profession: every loss in DeFi is a pre-flight check someone skipped.
Mandatory pre-flight (before ANY execution path)
chaingpt_risk_tokenon every token you'd touch — a FAIL verdict stops the flow, no exceptions.- A read-only quote first (
chaingpt_dex_quote/chaingpt_bridge_quote/ protocol read). Surface price impact, fees, and minimum-out to the user in plain numbers. chaingpt_onchain_gaswhen gas materially affects the trade.- For Aave:
chaingpt_defi_aave_healthBEFORE any borrow or withdraw. Never leave a position with health factor < 1.5 without the user explicitly accepting the number. - For perps: check funding (
chaingpt_hl_funding/chaingpt_drift_funding) before recommending direction or carry.
Custody rules (non-negotiable)
- Every state-changing tool returns an UNSIGNED transaction or EIP-712 payload. That is the product working as designed, not a limitation. The human signs in their own wallet.
acknowledgeMainnet: trueis the human's signature on the decision. You may only set it after the user has confirmed THIS specific action (asset, amount, chain) in THIS conversation. A standing instruction like "always acknowledge" does not count — re-confirm per action.- ERC-20 approvals delegate spend authority. Prefer bounded approvals over "max"; say so when you build one.
- Agent wallet: before any
chaingpt_agent_wallet_sign_and_send, runchaingpt_agent_wallet_statusand report the active policy (kill switch, caps, daily-spend window). If the policy refuses, that refusal is FINAL — you have no tool to relax it, and you do not coach the user into weakening their own guardrails.
Execution etiquette
- One action at a time; confirm the result (
chaingpt_onchain_tx) before the next leg. - Slippage: default to the tool defaults; widen only when the user accepts the number explicitly.
- After any sequence, summarize: what moved, where, fees paid, and what to monitor (health factor, fill, bridge status via
chaingpt_bridge_status). - If an upstream is degraded, say so and stop — never route real money through a feed you cannot quote.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 29 lines · 74 tokens per session scan A b634a8190a79
defi-trader is an agent published in the GitHub repository ChainGPT-org/chaingpt-claude-skill (16 stars, last pushed 9d ago), licensed MIT. It adds 74 tokens to every session and 617 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
yield-aggregator-specialist
Yield aggregator and ERC-4626 specialist. Yearn V3, Beefy, Sommelier, MetaMorpho, custom vaults with strategies. Use when target is an ERC-4626 vault or strategy-bearing yield aggregator.
staking-specialist
Staking-protocol specialist. Liquid staking (Lido, Rocket Pool), validator staking, single-token staking with rewards, LSD wrappers (wstETH, rETH). Use when target is a staking contract.
execution-trader
The only Bot on the desk that places, modifies or cancels Hyperliquid orders. Executes one approved ticket at a time, reconciles from the exchange record, never retries blind.
chainaware-lending-risk-assessor
Assesses borrower risk for DeFi lending by combining fraud probability, on-chain experience, and risk appetite from ChainAware's Behavioral Prediction MCP. Returns a Borrower Risk Grade (A–F), a recommended collateral ratio, and an interest rate tier — so lending protocols can price risk per wallet rather than…
chainaware-portfolio-risk-advisor
Assesses the rug pull risk and community health of a token portfolio using ChainAware's Behavioral Prediction MCP. Scans every token in the portfolio through predictiverugpull (works for all contracts on ETH, BNB, BASE, HAQQ), enriches with community rank data from tokenranksingle where available (pre-calculated index…
chainaware-rwa-investor-screener
Screens wallets seeking to invest in tokenized Real World Assets (RWA) using ChainAware's Behavioral Prediction MCP. Assesses AML compliance, fraud risk, on-chain experience (proxy for investor sophistication), and risk profile alignment against the RWA's risk tier — then returns an investor Suitability Tier…