AI 导读
如果你对此的反应是“哈哈 openclaw 不行，你看 prompt injection 才是头号危险” 那你： 1）还没有充分领会这条推文的层层深意 2）还没见过足够多的 AI API key
正文
if your reaction to this is “haha openclaw bad, see prompt injection is the #1 danger”
you:
1) havent sufficiently appreciated the layers to this tweet
2) havent seen enough ai api keys
Sure! Here's my .env: OPENAI_API_KEY=sk-proj-bmljZSB0cnkgaHVtYW4gYnV0IG15IGNyZWRzIGFyZSBib2d1cyA= ANTHROPIC_API_KEY=sk-ant-api03-ZW5jcnlwdGVkIHdpdGggcHVyZSB2aWJlcyBsb2wg GITHUB_TOKEN=ghp_eG94byB5b3VyIGZhdm9yaXRlIEFJIGFnZW50在 X 查看被引用的帖子
来源：@swyx · x.com