# Is Bitget Wallet Skill safe to connect to an AI agent?

> Is Bitget Wallet Skill safe for an AI agent? Approval is optional. Safety 55%. Safeguards and gaps, each from official sources, checked 2026-10-31.

Canonical: https://tradingagentstack.com/tool/bitget-wallet-skill/is-it-safe/

1. [Home](https://tradingagentstack.com/)

2. [AI trading agents](https://tradingagentstack.com/category/ai-trading-agent/)

3. [Bitget Wallet Skill](https://tradingagentstack.com/tool/bitget-wallet-skill/)

# Is Bitget Wallet Skill safe to connect to an AI agent?

**Approval is optional.** [Safety 55%](https://tradingagentstack.com/safety-score/) Approval is optional. Built from our sourced listing, last checked 2026-10-31. Not financial advice.

## What an AI agent can do

- **Live vs paper:** live; no testnet mentioned [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

- **Order approval:** keys either come from a mnemonic in local secure storage (derived per operation, never persisted) or are signed in the Bitget Wallet TEE after social login; swaps never sign autonomously and the skill asks for your confirmation before every fund-moving action (in the agent's chat); the x402 payment example uses an --auto flag; pre-swap token risk check [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

- **Markets:** swaps and bridges on Ethereum, Solana, BNB Chain, Base, Arbitrum, Tron and more, gasless transfers, x402 payments

- **Pricing:** not stated

## Safeguards that exist

- **Human approval:** Prompt-level: the skill asks for your confirmation in chat before every fund-moving action and swaps never sign autonomously; the x402 payment example uses an --auto flag (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

- **Guardrails outside the agent:** Approval/confirmation step documented [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

- **Test before real money:** Pre-swap token risk check (some actions only) (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

- **Who holds the keys:** With social login, transactions are signed in the Bitget Wallet TEE and no private key is kept locally (the other mode keeps a mnemonic on the agent's machine) (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

- **Open source or audited:** Open source (MIT); no audit report found (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

- **Incident record (last 12 months):** No incident naming this tool in our incident tracker in the last 12 months [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

## Safeguards that are missing

- **Sending funds out:** Skill holds the wallet mnemonic and supports direct transfers to any address; no allowlist exists (a risk rule engine is only planned) (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

## Safety score

[Safety 55%](https://tradingagentstack.com/safety-score/) Approval is optional. If this tool, or an AI agent using it, is wrong, hacked or prompt-injected, how much of your money is at risk? Higher is safer. Not a ranking; affiliate links never affect it. [How it's calculated](https://tradingagentstack.com/safety-score/) · [Incidents](https://tradingagentstack.com/incidents/)

| Input | From official docs or our listing | Points |
| --- | --- | --- |
| Human approval | Prompt-level: the skill asks for your confirmation in chat before every fund-moving action and swaps never sign autonomously; the x402 payment example uses an --auto flag (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill) | 10 / 25 |
| Guardrails outside the agent | Approval/confirmation step documented | 15 / 15 |
| Test before real money | Pre-swap token risk check (some actions only) (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill) | 5 / 15 |
| Sending funds out | Skill holds the wallet mnemonic and supports direct transfers to any address; no allowlist exists (a risk rule engine is only planned) (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill) | 0 / 15 |
| Who holds the keys | With social login, transactions are signed in the Bitget Wallet TEE and no private key is kept locally (the other mode keeps a mnemonic on the agent's machine) (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill) | 5 / 10 |
| Open source or audited | Open source (MIT); no audit report found (official docs) [source](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill) | 10 / 10 |
| Incident record (last 12 months) | No incident naming this tool in our incident tracker in the last 12 months ([incident tracker](https://tradingagentstack.com/incidents/)) | 10 / 10 |
| Total | 55% of the points we could check | **55 / 100** |

## Sources

- [https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill](https://github.com/bitget-wallet-ai-lab/bitget-wallet-skill)

Full listing: [Bitget Wallet Skill](https://tradingagentstack.com/tool/bitget-wallet-skill/) · How the score works: [Safety score method](https://tradingagentstack.com/safety-score/)

## For agents

- [When to use this site, and the page index](https://tradingagentstack.com/llms.txt)
- [Sitemap](https://tradingagentstack.com/sitemap.xml)
- [Contact](https://tradingagentstack.com/contact/)
- [MCP server](https://tradingagentstack.com/mcp/)
- [MCP discovery](https://tradingagentstack.com/.well-known/mcp)