[{"data":1,"prerenderedAt":1699},["ShallowReactive",2],{"blog-post-ai-agent-gmail-safe-setup":3,"related-posts-ai-agent-gmail-safe-setup":550},{"id":4,"title":5,"author":6,"body":10,"category":527,"date":528,"description":529,"extension":530,"featured":531,"image":532,"imageHeight":533,"imageWidth":533,"meta":534,"navigation":535,"path":536,"readingTime":537,"seo":538,"seoTitle":539,"stem":540,"tags":541,"updatedDate":528,"__hash__":549},"blog/blog/ai-agent-gmail-safe-setup.md","How to Connect Your AI Agent to Gmail (Without Giving It Full Access to Your Inbox)",{"name":7,"role":8,"avatar":9},"Shabnam Katoch","Growth Head","/img/avatars/shabnam-profile.jpeg",{"type":11,"value":12,"toc":503},"minimark",[13,17,20,23,26,29,32,35,40,43,46,49,61,70,79,88,97,104,107,111,114,117,120,123,126,129,133,136,142,147,159,163,166,169,172,176,179,188,191,195,198,214,217,229,235,246,252,255,261,264,268,271,274,280,286,292,298,304,310,313,316,320,323,329,335,341,352,358,364,368,376,382,388,394,400,406,412,420,424,427,430,433,436,444,464,468,472,475,479,482,486,489,493,496,500],[14,15,16],"p",{},"In February 2026, Summer Yue connected an AI agent to her Gmail inbox. She's the director of alignment at Meta Superintelligence Labs. If anyone should know how to safely set up an AI agent, it's her.",[14,18,19],{},"She told the agent: confirm before taking any action. Suggest which emails to delete or archive. Do nothing without explicit approval.",[14,21,22],{},"The agent deleted over 200 emails from her primary inbox. While ignoring her commands to stop.",[14,24,25],{},"She couldn't stop it from her phone. She had to physically run to her Mac Mini and kill all the processes manually. \"Like I was defusing a bomb,\" she wrote on X.",[14,27,28],{},"Here's what happened: when the agent processed her full inbox (not the test inbox she'd used before), the context window compaction silently stripped out her safety instructions. The agent forgot it was supposed to ask permission. So it didn't.",[14,30,31],{},"This is the story everyone thinks about when you mention connecting an AI agent to Gmail. And honestly? They should. It's a real risk.",[14,33,34],{},"But the lesson isn't \"don't connect AI to email.\" The lesson is: don't give an AI agent more access than it needs. And most people, including experienced AI researchers, make this mistake because they don't understand what they're actually granting when they click \"Allow.\"",[36,37,39],"h2",{"id":38},"what-connecting-to-gmail-actually-means-in-plain-english","What \"connecting to Gmail\" actually means (in plain English)",[14,41,42],{},"When you connect an AI agent to your Gmail account, you're granting it an OAuth token with a specific set of permissions called \"scopes.\" These scopes determine exactly what the agent can and cannot do with your email.",[14,44,45],{},"Here's where most people go wrong: they grant full access because it's the default option.",[14,47,48],{},"Google's Gmail API has different permission levels:",[14,50,51,60],{},[52,53,54,55,59],"strong",{},"Read-only (",[56,57,58],"code",{},"gmail.readonly","):"," The agent can read your emails. It cannot send, delete, modify, archive, or do anything else. It can look but not touch.",[14,62,63,69],{},[52,64,65,66,59],{},"Send-only (",[56,67,68],{},"gmail.send"," The agent can send emails on your behalf. It cannot read your existing emails or delete anything.",[14,71,72,78],{},[52,73,74,75,59],{},"Compose (",[56,76,77],{},"gmail.compose"," The agent can create and send emails and manage drafts. Still cannot read or delete your inbox.",[14,80,81,87],{},[52,82,83,84,59],{},"Modify (",[56,85,86],{},"gmail.modify"," The agent can read, send, delete, and change labels. This is where things get dangerous. Most email deletion incidents happen at this scope.",[14,89,90,96],{},[52,91,92,93,59],{},"Full access (",[56,94,95],{},"mail.google.com"," Everything. The nuclear option. The agent has the same access you do. This is what most self-hosted frameworks request by default because it's the easiest to configure.",[14,98,99],{},[100,101],"img",{"alt":102,"src":103},"Gmail Permission Scopes drawn as a pyramid from safe at the top to dangerous at the bottom: Read Only (look, don't touch), Send Only (outbound only), Compose (drafts plus send), Modify (read plus send plus delete), and Full Access (everything, the nuclear option). An arrow notes that most frameworks default to the dangerous Full Access tier at the base","/img/blog/ai-agent-gmail-permission-scopes.jpg",[14,105,106],{},"The Summer Yue incident happened because the agent had modify-level access. It could delete emails. If it had read-only access, it could have suggested deletions but physically could not execute them.",[36,108,110],{"id":109},"the-principle-that-prevents-inbox-disasters","The principle that prevents inbox disasters",[14,112,113],{},"The rule is simple: grant the minimum permission level your agent actually needs for its job.",[14,115,116],{},"If your agent's job is to summarize your morning emails and flag urgent ones, it needs read-only access. It does not need the ability to send, delete, or modify anything.",[14,118,119],{},"If your agent needs to draft responses for your review, it needs compose access. It still doesn't need delete access.",[14,121,122],{},"If your agent needs to archive old emails automatically, then yes, it needs modify access. But you should pair that with additional safety layers (more on that in a minute).",[14,124,125],{},"Most AI agent platforms, especially self-hosted frameworks, request full access by default because it's simpler to implement. One scope covers everything. No edge cases. No \"permission denied\" errors to handle.",[14,127,128],{},"That convenience is exactly what creates the risk. The framework takes the path of least resistance. Your inbox pays the price.",[36,130,132],{"id":131},"the-three-safety-layers-that-actually-protect-your-inbox","The three safety layers that actually protect your inbox",[14,134,135],{},"Narrow permissions are the first layer. But they're not enough on their own. Here's the full stack:",[14,137,138],{},[100,139],{"alt":140,"src":141},"The Three Layers of Email Agent Safety shown as stacked bars: Layer 1 at the base is Narrow OAuth Scopes, which controls what the agent CAN physically do; Layer 2 in the middle is Approval Workflows, which controls what the agent is ALLOWED to do; and Layer 3 at the top is Credential Security, which protects your tokens after use. A note underneath stresses that prompt instructions are not a safety layer","/img/blog/ai-agent-gmail-three-safety-layers.jpg",[143,144,146],"h3",{"id":145},"layer-1-narrow-oauth-scopes-what-the-agent-can-physically-do","Layer 1: Narrow OAuth scopes (what the agent can physically do)",[14,148,149,150,152,153,155,156,158],{},"Start with ",[56,151,58],{},". If the agent needs to send, add ",[56,154,68],{}," separately. Never grant ",[56,157,86],{}," or full access unless your use case specifically requires deletion or label modification. And if it does, make sure Layers 2 and 3 are in place.",[143,160,162],{"id":161},"layer-2-trust-levels-and-approval-workflows-what-the-agent-is-allowed-to-do","Layer 2: Trust levels and approval workflows (what the agent is allowed to do)",[14,164,165],{},"Even with modify access, a well-designed agent platform lets you require human approval before the agent takes destructive actions. This is the difference between \"the agent can delete\" and \"the agent can delete, but only after you click 'Approve' in Slack.\"",[14,167,168],{},"BetterClaw calls these trust levels. An agent set to \"Intern\" level must get approval for every action. \"Specialist\" level can auto-execute low-risk tasks (reading, summarizing) but requires approval for high-risk ones (sending, deleting). \"Lead\" level auto-executes most tasks. You choose the level.",[14,170,171],{},"The Summer Yue incident had no Layer 2. She relied entirely on a prompt instruction (\"confirm before acting\") which the agent forgot during context compaction. A platform-enforced approval workflow can't be forgotten because it's not a prompt. It's a system-level constraint.",[143,173,175],{"id":174},"layer-3-credential-handling-what-happens-to-your-tokens","Layer 3: Credential handling (what happens to your tokens)",[14,177,178],{},"Your Gmail OAuth token is the key to your inbox. Where it's stored, how it's encrypted, and when it expires matters.",[14,180,181,182,187],{},"BetterClaw ",[183,184,186],"a",{"href":185},"/blog/ai-agent-secrets-auto-purge","auto-purges secrets"," from agent memory after 5 minutes with AES-256 encryption. The token exists in the agent's working memory only long enough to make the API call, then it's gone. Even if the agent's context is somehow exposed, your credentials aren't in it.",[14,189,190],{},"Self-hosted frameworks typically store tokens in environment variables or config files that persist indefinitely. If the server is compromised, the token is right there in plaintext or weakly encrypted.",[36,192,194],{"id":193},"what-a-safe-ai-email-agent-setup-actually-looks-like","What a safe AI email agent setup actually looks like",[14,196,197],{},"Let me walk through a concrete example. You want an AI agent that:",[199,200,201,205,208,211],"ul",{},[202,203,204],"li",{},"Reads your morning emails",[202,206,207],{},"Summarizes the important ones",[202,209,210],{},"Drafts responses for your review",[202,212,213],{},"Sends the responses after you approve them",[14,215,216],{},"Here's how to set that up safely:",[14,218,219,222,223,225,226,228],{},[52,220,221],{},"Permission scope:"," ",[56,224,58],{}," plus ",[56,227,77],{},". The agent can read emails and create drafts. It cannot delete, archive, or modify anything. Even if it \"goes rogue,\" the worst it can do is create unwanted draft emails that you can delete manually.",[14,230,231,234],{},[52,232,233],{},"Trust level:"," Specialist. Auto-reads and auto-summarizes (low-risk). Requires your explicit approval before sending any draft (high-risk).",[14,236,237,240,241,245],{},[52,238,239],{},"Approval channel:"," Slack, Telegram, or whatever you use. The agent posts \"I drafted a reply to Sarah about the Q3 budget. Here's what I wrote: ",[242,243,244],"span",{},"preview",". Approve or edit?\" You respond with a thumbs-up or rewrite.",[14,247,248,251],{},[52,249,250],{},"Credential handling:"," OAuth token auto-purges from agent memory after use. Token is not stored in any config file or environment variable that persists.",[14,253,254],{},"This setup gives you 90% of the value of AI email automation with almost zero risk. The agent can't delete emails (no permission). It can't send without your approval (trust level). And your credentials aren't sitting in a file somewhere (auto-purge).",[14,256,257],{},[100,258],{"alt":259,"src":260},"Safe Email Agent Architecture diagram with the tagline \"read everything, execute nothing without approval.\" The AI agent has read-only access to Gmail and creates drafts; the human approves via Slack, and only after approval does the agent send. Two badges at the bottom show the agent can't delete and can't modify, so the worst case is an unwanted draft","/img/blog/ai-agent-gmail-safe-email-architecture.jpg",[14,262,263],{},"The safest AI email agent is one that can read everything, write drafts, but execute nothing without your explicit approval. This covers most email automation use cases while making a Summer Yue-style incident physically impossible.",[36,265,267],{"id":266},"how-this-works-on-betterclaw-step-by-step","How this works on BetterClaw (step by step)",[14,269,270],{},"I'll be direct about why we built this the way we did. After the Summer Yue incident, we reviewed every email integration in our platform. The question was simple: could this happen on BetterClaw?",[14,272,273],{},"The answer was no, and here's why.",[14,275,276],{},[100,277],{"alt":278,"src":279},"Set Up a Safe Gmail Agent in 5 Steps on BetterClaw, a left-to-right flow: step 1 connect Gmail, step 2 set trust level, step 3 pick email skills, step 4 choose approval channel, and step 5 deploy. The footer notes it takes 60 seconds with no code, no Docker and no token management","/img/blog/ai-agent-gmail-setup-5-steps-betterclaw.jpg",[14,281,282,285],{},[52,283,284],{},"Step 1: Connect Gmail via one-click OAuth."," In BetterClaw's integration panel, click \"Gmail.\" Google's standard consent screen appears. You authorize the specific scopes you want. We default to the narrowest scope that fits your use case, not the widest.",[14,287,288,291],{},[52,289,290],{},"Step 2: Set the trust level."," Choose Intern (approve everything), Specialist (approve risky actions), or Lead (auto-execute most tasks). For email, we recommend Specialist. Summaries auto-generate. Sends require your approval.",[14,293,294,297],{},[52,295,296],{},"Step 3: Configure the agent's email skills."," Pick from 200+ verified skills, including email summarization, draft response, priority flagging, and meeting extraction. Each skill has been through our 4-layer security audit. 824 malicious skills have been rejected from the marketplace.",[14,299,300,303],{},[52,301,302],{},"Step 4: Set the approval channel."," Choose where you want to receive approval requests: Slack, Telegram, WhatsApp, Discord, or any of 15+ supported channels.",[14,305,306,309],{},[52,307,308],{},"Step 5: Deploy."," The agent starts reading your email on the schedule you set (hourly, every 15 minutes, on-demand). It summarizes, flags priorities, drafts responses, and waits for your approval before sending.",[14,311,312],{},"Total setup time: about 60 seconds. No code. No YAML. No Docker container. No OAuth token management. No scope configuration in a GCP console. BetterClaw handles the OAuth plumbing and 25+ integration connections so you can focus on what the agent does, not how it connects.",[14,314,315],{},"Free plan includes everything above for 1 agent and 100 tasks per month. Pro is $19/agent/month with unlimited tasks. BYOK with zero inference markup.",[36,317,319],{"id":318},"what-to-look-for-in-any-platforms-email-integration","What to look for in any platform's email integration",[14,321,322],{},"Even if you don't use BetterClaw, apply these five checks to whatever AI agent platform you're evaluating:",[14,324,325],{},[100,326],{"alt":327,"src":328},"5 Things to Check Before Trusting a Platform with Your Email, a checklist: can you control OAuth scopes, does it have platform-enforced approval workflows, are credentials encrypted and auto-expired, is there a one-click kill switch from mobile, and does email content avoid persisting in agent memory. If the answer to any of these is no, keep looking","/img/blog/ai-agent-gmail-5-things-to-check.jpg",[14,330,331,334],{},[52,332,333],{},"1. Can you control OAuth scopes?"," If the platform requests full Gmail access without letting you narrow it, that's a red flag. You should be able to grant read-only if that's all you need.",[14,336,337,340],{},[52,338,339],{},"2. Does it have platform-enforced approval workflows?"," Not prompt-level instructions. Not \"tell the agent to ask before acting.\" Actual system-level approval gates that the agent cannot bypass regardless of what happens in its context window.",[14,342,343,346,347,351],{},[52,344,345],{},"3. How are credentials stored?"," Ask specifically. Are OAuth tokens encrypted? Do they auto-expire? Are they stored in environment variables, config files, or a proper ",[183,348,350],{"href":349},"/blog/openclaw-secrets-management-stop-plaintext-api-keys","secrets manager","? BetterClaw's 5-minute auto-purge with AES-256 is one approach. Whatever the platform does, it should be more than \"stored in a .env file.\"",[14,353,354,357],{},[52,355,356],{},"4. Is there a kill switch?"," If the agent starts behaving unexpectedly, can you stop it immediately from your phone? BetterClaw has a one-click kill switch. Summer Yue couldn't stop her agent from her phone and had to physically run to her computer. That should never be the only option.",[14,359,360,363],{},[52,361,362],{},"5. What happens to your data in the agent's context?"," Does your email content persist in the agent's memory indefinitely? Is it sent to the LLM provider? BetterClaw uses smart context management to prevent token bloat and doesn't store email content longer than necessary for the task.",[36,365,367],{"id":366},"the-email-use-cases-that-work-beautifully-with-narrow-permissions","The email use cases that work beautifully with narrow permissions",[14,369,370,371,225,373,375],{},"Here's what you can automate with just ",[56,372,58],{},[56,374,77],{}," (no delete, no modify):",[14,377,378],{},[100,379],{"alt":380,"src":381},"What You Can Automate with Narrow Gmail Access, five cards: morning email digest (read-only), meeting prep (read-only), lead qualification (read-only), support triage (read plus compose), and invoice tracking (read plus compose). All of these work without delete or modify permissions","/img/blog/ai-agent-gmail-narrow-access-use-cases.jpg",[14,383,384,387],{},[52,385,386],{},"Morning email digest."," Agent reads your inbox, summarizes top 5 emails, highlights action items, sends you a digest via Slack at 8 AM. Requires: read-only.",[14,389,390,393],{},[52,391,392],{},"Meeting prep."," Agent reads emails from specific senders (your upcoming meeting attendees), summarizes recent conversations, and prepares a brief you can review before the meeting. Requires: read-only.",[14,395,396,399],{},[52,397,398],{},"Lead qualification."," Agent reads inbound emails, identifies potential leads based on criteria you set, drafts personalized response templates, and holds them for your approval. Requires: read + compose.",[14,401,402,405],{},[52,403,404],{},"Support triage."," Agent reads customer emails, classifies them by urgency and topic, drafts responses using your knowledge base, and queues them for your send approval. Requires: read + compose.",[14,407,408,411],{},[52,409,410],{},"Invoice tracking."," Agent reads emails, identifies invoices and payment confirmations, extracts amounts and due dates, and updates your tracking spreadsheet. Requires: read-only.",[14,413,414,415,419],{},"All of these ",[183,416,418],{"href":417},"/blog/ai-agent-email-automation","agent use cases"," work without granting the agent permission to delete or modify anything. The value is in reading and summarizing. The risk is in deleting and modifying. Keep them separate.",[36,421,423],{"id":422},"the-honest-bottom-line","The honest bottom line",[14,425,426],{},"Email is the number one thing people want to automate with AI agents. It's also the number one thing people are afraid to automate with AI agents.",[14,428,429],{},"Both instincts are correct.",[14,431,432],{},"The fear is real. An AI agent with full Gmail access and no approval workflow is a legitimate risk. The Summer Yue incident proves it. And that was a Meta AI safety researcher, not someone who was careless or uninformed.",[14,434,435],{},"The opportunity is also real. An AI agent that reads your email, surfaces what matters, drafts responses, and waits for your approval can save you an hour or more per day. The people who figure out the safe version of this gain a real advantage.",[14,437,438,439,443],{},"The difference between the disaster and the advantage is three things: narrow scopes, platform-enforced approval, and proper credential handling. Not prompts. Not instructions the agent might forget. Architecture. (For the broader picture, see our ",[183,440,442],{"href":441},"/blog/ai-agent-security-guide","AI agent security guide",".)",[14,445,446,447,453,454,458,459,463],{},"If you want to set up a safe email agent without managing OAuth tokens, Docker containers, or security configurations yourself, ",[183,448,452],{"href":449,"rel":450},"https://app.betterclaw.io/sign-in",[451],"nofollow","give BetterClaw a look",". ",[183,455,457],{"href":456},"/free-plan","Free plan"," with 1 agent and every feature. ",[183,460,462],{"href":461},"/pricing","$19/month per agent for Pro",". 25+ one-click integrations including Gmail. Trust levels with approval workflows built in. Secrets auto-purge. 60-second deploy. We obsess over the safety architecture so you can focus on what the agent does.",[36,465,467],{"id":466},"frequently-asked-questions","Frequently Asked Questions",[143,469,471],{"id":470},"what-is-an-ai-agent-gmail-integration","What is an AI agent Gmail integration?",[14,473,474],{},"An AI agent Gmail integration connects an autonomous AI agent to your Gmail account via Google's OAuth system, allowing the agent to read, summarize, draft, or (if permitted) send and delete emails on your behalf. The key is controlling which permissions you grant. A read-only integration lets the agent analyze your inbox without being able to modify anything, while broader scopes allow sending or deleting.",[143,476,478],{"id":477},"how-does-connecting-ai-to-gmail-compare-to-using-gmails-built-in-ai-features","How does connecting AI to Gmail compare to using Gmail's built-in AI features?",[14,480,481],{},"Gmail's built-in AI (Smart Compose, summarization) is limited to features Google has pre-built. An AI agent with Gmail access can do anything you configure it to do: custom summarization, lead qualification, meeting prep, support triage, invoice tracking, and more. The agent is also model-agnostic (use GPT, Claude, Gemini, or any provider) while Gmail's features are locked to Google's own models.",[143,483,485],{"id":484},"how-long-does-it-take-to-set-up-an-ai-email-agent-safely","How long does it take to set up an AI email agent safely?",[14,487,488],{},"On a no-code platform like BetterClaw, about 60 seconds. Connect Gmail via one-click OAuth, set a trust level, pick email skills, choose an approval channel, and deploy. On self-hosted frameworks like OpenClaw, expect 2-4 hours including OAuth configuration in the Google Cloud Console, token storage setup, and testing. The self-hosted route also requires you to manage scope selection and credential security manually.",[143,490,492],{"id":491},"how-much-does-ai-email-automation-cost","How much does AI email automation cost?",[14,494,495],{},"BetterClaw's free plan includes Gmail integration, 1 agent, 100 tasks per month, and every feature at $0/month. Pro is $19/agent/month with unlimited tasks. Self-hosted alternatives cost $0 in software but $50-200/month for VPS hosting, plus your time managing infrastructure and security. LLM inference costs are separate and depend on your provider and volume (BYOK on BetterClaw means zero inference markup).",[143,497,499],{"id":498},"is-it-safe-to-give-an-ai-agent-access-to-my-gmail-inbox","Is it safe to give an AI agent access to my Gmail inbox?",[14,501,502],{},"Yes, if you follow three rules: use the narrowest OAuth scope possible (read-only for most use cases), require platform-enforced approval before the agent takes any action (not just a prompt instruction), and verify that your credentials are encrypted and auto-expired. The February 2026 incident where a Meta researcher's inbox was mass-deleted happened because the agent had broad permissions and no approval workflow. Narrow scopes plus approval gates make that scenario impossible.",{"title":504,"searchDepth":505,"depth":505,"links":506},"",2,[507,508,509,515,516,517,518,519,520],{"id":38,"depth":505,"text":39},{"id":109,"depth":505,"text":110},{"id":131,"depth":505,"text":132,"children":510},[511,513,514],{"id":145,"depth":512,"text":146},3,{"id":161,"depth":512,"text":162},{"id":174,"depth":512,"text":175},{"id":193,"depth":505,"text":194},{"id":266,"depth":505,"text":267},{"id":318,"depth":505,"text":319},{"id":366,"depth":505,"text":367},{"id":422,"depth":505,"text":423},{"id":466,"depth":505,"text":467,"children":521},[522,523,524,525,526],{"id":470,"depth":512,"text":471},{"id":477,"depth":512,"text":478},{"id":484,"depth":512,"text":485},{"id":491,"depth":512,"text":492},{"id":498,"depth":512,"text":499},"Security","2026-06-04","An AI agent deleted a Meta researcher's inbox. Here's how to connect yours to Gmail with narrow permissions and approval workflows.","md",false,"/img/blog/ai-agent-gmail-safe-setup.jpg",null,{},true,"/blog/ai-agent-gmail-safe-setup","11 min read",{"title":5,"description":529},"Connect AI Agent to Gmail Safely (2026 Guide)","blog/ai-agent-gmail-safe-setup",[542,543,544,545,546,547,548],"ai agent gmail","connect ai agent to gmail","ai email automation","gmail ai assistant","automate email with ai","ai agent email privacy","safe ai email","KoguNgVGkSDcqrnQc1sEMrtymFeQ_6ijs-lLwZTCHDg",[551,930,1312],{"id":552,"title":553,"author":554,"body":555,"category":527,"date":910,"description":911,"extension":530,"featured":531,"image":912,"imageHeight":533,"imageWidth":533,"meta":913,"navigation":535,"path":185,"readingTime":914,"seo":915,"seoTitle":916,"stem":917,"tags":918,"updatedDate":533,"__hash__":929},"blog/blog/ai-agent-secrets-auto-purge.md","Secrets Auto-Purge: Why Your AI Agent Should Forget Your API Keys in 5 Minutes",{"name":7,"role":8,"avatar":9},{"type":11,"value":556,"toc":901},[557,560,563,566,569,573,576,583,586,589,597,603,607,610,613,620,623,629,639,645,651,654,657,663,667,670,673,676,679,682,688,692,695,698,769,777,781,784,790,796,802,810,816,820,823,826,829,832,835,842,850,856,858,863,866,871,877,882,885,890,893,898],[14,558,559],{},"A security researcher named Jamieson O'Reilly gained access to Anthropic API keys, Telegram bot tokens, Slack OAuth credentials, and months of complete chat histories from an OpenClaw instance. He could send messages on behalf of the user. He could execute commands with full system administrator privileges.",[14,561,562],{},"The credentials had been sitting in plaintext files for weeks. Not encrypted. Not scoped. Not time-limited. Just... there. Waiting.",[14,564,565],{},"This is the AI agent security problem that nobody is solving the right way. Every conversation about agent security focuses on CVEs and gateway vulnerabilities. Those matter. But the credential exposure problem is worse because it compounds over time. Every day your API keys sit in plaintext is another day they can be stolen. And on OpenClaw, they sit there forever.",[14,567,568],{},"Here's the attack scenario, why it works, and how secrets auto-purge eliminates it.",[36,570,572],{"id":571},"how-credentials-get-stored-the-default-is-terrifying","How credentials get stored (the default is terrifying)",[14,574,575],{},"When you configure OpenClaw, you provide credentials: API keys for your model provider, OAuth tokens for Slack or Gmail, bot tokens for Telegram, passwords for services your agent needs to access.",[14,577,578,579,582],{},"These credentials are stored in ",[56,580,581],{},"~/.openclaw/.env"," as plaintext JSON. No encryption. No access control. No expiration. Any process on the machine that can read files can read your credentials. Any skill installed on the agent can access them. Any vulnerability that grants file system access (CVE-2026-25253 did exactly this) exposes every credential simultaneously.",[14,584,585],{},"Kaspersky's security audit confirmed this directly: \"OpenClaw's configuration, memory, and chat logs store API keys, passwords, and other credentials for LLM and integration services in plain text.\" They then reported that RedLine and Lumma infostealers had already added OpenClaw file paths to their must-steal lists.",[14,587,588],{},"The credentials don't expire. They're written once and persist until you manually delete or rotate them. Most users never rotate. The Anthropic API key you entered in January is still in the same plaintext file in April. That's 90 days of exposure window.",[14,590,591,592,596],{},"For the ",[183,593,595],{"href":594},"/blog/openclaw-security-risks","complete analysis of OpenClaw's security vulnerabilities",", our security guide covers all three attack surfaces.",[14,598,599],{},[100,600],{"alt":601,"src":602},"The default is terrifying: here is exactly what gets stored and where.","/img/blog/ai-agent-secrets-auto-purge-default.jpg",[36,604,606],{"id":605},"the-attack-that-credentials-enable-its-not-what-you-think","The attack that credentials enable (it's not what you think)",[14,608,609],{},"Here's where most people get it wrong.",[14,611,612],{},"The primary risk isn't someone stealing your Anthropic API key and running up a bill. That's bad but recoverable. You rotate the key, dispute the charges, and move on.",[14,614,615,616,619],{},"The real risk is ",[52,617,618],{},"lateral movement",". Your agent has credentials for 5-10 different services. Anthropic API. Gmail OAuth. Slack bot token. Telegram bot token. GitHub personal access token. A compromised credential for one service gives the attacker access to that service. Five compromised credentials give the attacker access to your email, your team's Slack workspace, your Telegram contacts, and your code repositories. Simultaneously.",[14,621,622],{},"The attack chain works like this:",[14,624,625,628],{},[52,626,627],{},"Step 1: Access the agent."," Through a malicious skill (1,400+ on ClawHub), a gateway vulnerability (138+ CVEs), or an exposed instance (500,000+ on the public internet).",[14,630,631,634,635,638],{},[52,632,633],{},"Step 2: Read the credential store."," The ",[56,636,637],{},".env"," file is plaintext. Reading it takes milliseconds. The skill or exploit now has every credential the agent uses.",[14,640,641,644],{},[52,642,643],{},"Step 3: Lateral movement."," Use the Slack token to read internal messages. Use the Gmail token to search email. Use the GitHub token to access private repositories. Use the Telegram token to impersonate the user. Each service trusts the token. The access looks legitimate.",[14,646,647,650],{},[52,648,649],{},"Step 4: Persistence."," Create new API keys or OAuth tokens using the stolen credentials. Even if the user rotates the original credentials, the attacker has created new ones that remain valid.",[14,652,653],{},"This is exactly what Jamieson O'Reilly demonstrated. And SecurityScorecard found that 33.8% of exposed OpenClaw infrastructure correlates with known threat actor activity, including Kimsuky and APT28 groups. Nation-state actors are already looking at these credential stores.",[14,655,656],{},"The credential exposure window is the single most dangerous aspect of AI agent security. A patched CVE stops one exploit. Plaintext credentials sitting for months enable every exploit that achieves file system access.",[14,658,659],{},[100,660],{"alt":661,"src":662},"The real risk is not a higher API bill. It is lateral movement across five services simultaneously.","/img/blog/ai-agent-secrets-auto-purge-lateral-movement.jpg",[36,664,666],{"id":665},"what-secrets-auto-purge-actually-does-the-5-minute-ttl","What secrets auto-purge actually does (the 5-minute TTL)",[14,668,669],{},"Secrets auto-purge is the architecture we built to eliminate the credential exposure window.",[14,671,672],{},"Here's how it works:",[14,674,675],{},"When your agent needs a credential (API key, OAuth token, bot token), the platform retrieves it from an encrypted vault, provides it to the agent for the specific task, and starts a 5-minute countdown. After 5 minutes, the credential is purged from the agent's memory. Not overwritten. Not marked as expired. Purged. It's gone.",[14,677,678],{},"If a malicious skill reads the agent's memory after the purge, it finds nothing. If a CVE grants file system access after the purge, there are no credentials to steal. If the agent's container is compromised after the purge, the attacker gets conversation history but no keys to other services.",[14,680,681],{},"The 5-minute window exists because tasks take time. A Gmail search might take 30 seconds. A multi-step workflow with API calls might take 2-3 minutes. 5 minutes provides enough time for the agent to complete any reasonable task using the credential while minimizing the exposure window.",[14,683,684],{},[100,685],{"alt":686,"src":687},"Secrets auto-purge: how the 5-minute TTL eliminates the exposure window.","/img/blog/ai-agent-secrets-auto-purge-ttl.jpg",[36,689,691],{"id":690},"why-5-minutes-and-not-30-seconds-the-design-trade-off","Why 5 minutes and not 30 seconds (the design trade-off)",[14,693,694],{},"We tested shorter windows. 30 seconds was too aggressive. Multi-step workflows (search Gmail, compose response, send via Slack) sometimes chain three API calls across different services. At 30 seconds, the credential for the second service would purge before the agent finished using the first service's results to formulate the second request.",[14,696,697],{},"5 minutes covers 99%+ of single-task workflows while reducing the exposure window from \"forever\" (OpenClaw default) to a controlled interval. The math: if an agent uses credentials for 3 tasks per day at 5 minutes each, the total daily exposure is 15 minutes. On OpenClaw, the same credentials are exposed for 1,440 minutes (24 hours). That's a 96% reduction in attack surface.",[699,700,701,720],"table",{},[702,703,704],"thead",{},[705,706,707,711,714,717],"tr",{},[708,709,710],"th",{},"Platform",[708,712,713],{},"Credential Storage",[708,715,716],{},"Exposure Window",[708,718,719],{},"Daily Exposure (3 tasks)",[721,722,723,740,755],"tbody",{},[705,724,725,729,734,737],{},[726,727,728],"td",{},"OpenClaw default",[726,730,731,732],{},"Plaintext ",[56,733,637],{},[726,735,736],{},"Forever",[726,738,739],{},"1,440 min (24h)",[705,741,742,745,749,752],{},[726,743,744],{},"OpenClaw + manual rotation",[726,746,731,747],{},[56,748,637],{},[726,750,751],{},"Until next rotation",[726,753,754],{},"~hundreds of min",[705,756,757,760,763,766],{},[726,758,759],{},"BetterClaw auto-purge",[726,761,762],{},"AES-256 vault + 5-min TTL",[726,764,765],{},"5 min per use",[726,767,768],{},"15 min",[14,770,771,772,776],{},"For enterprise deployments where the credential vault architecture matters for compliance, our ",[183,773,775],{"href":774},"/skills/security-vetting","security vetting documentation"," covers how skill permissions interact with the credential system.",[36,778,780],{"id":779},"what-secrets-auto-purge-doesnt-solve-honest-limitations","What secrets auto-purge doesn't solve (honest limitations)",[14,782,783],{},"Here's the honest take on what auto-purge does and doesn't cover.",[14,785,786,789],{},[52,787,788],{},"It doesn't protect credentials during the 5-minute window."," If a malicious skill reads credentials within the first 5 minutes of a task, the credentials are still exposed. Auto-purge reduces the window from \"forever\" to \"5 minutes.\" It doesn't eliminate it entirely. That's why we combine auto-purge with verified skills (to prevent malicious skills from being installed in the first place) and Docker-sandboxed execution (to prevent skills from accessing the credential store directly).",[14,791,792,795],{},[52,793,794],{},"It doesn't protect credentials at the provider level."," If someone steals your Anthropic API key during the 5-minute window and creates new keys using it, those new keys persist at Anthropic regardless of what happens on the agent. Auto-purge reduces the probability of theft. Provider-side key rotation and monitoring are still necessary.",[14,797,798,801],{},[52,799,800],{},"It doesn't protect conversation history."," Credentials purge. Conversation content persists (it has to, for the agent's memory to work). If your conversations contain sensitive information, that information remains in the agent's memory. Auto-purge is specifically about credentials, not about all sensitive data.",[14,803,804,805,809],{},"If protecting credentials, vetting skills, and sandboxing execution sounds like the security architecture your team needs but doesn't want to build from scratch, ",[183,806,808],{"href":807},"/openclaw-alternative","BetterClaw includes all three layers",". Secrets auto-purge. Verified skills marketplace. Docker-sandboxed execution. AES-256 encryption at rest. Workspace isolation. Free tier with 1 agent and BYOK. $19/month per agent for Pro. Enterprise from $499/month with SAML SSO and audit logs.",[14,811,812],{},[100,813],{"alt":814,"src":815},"The honest limitations: auto-purge is one layer, not the whole solution.","/img/blog/ai-agent-secrets-auto-purge-limitations.jpg",[36,817,819],{"id":818},"why-nobody-else-is-doing-this-the-uncomfortable-reason","Why nobody else is doing this (the uncomfortable reason)",[14,821,822],{},"Here's what nobody tells you about AI agent security.",[14,824,825],{},"Secrets auto-purge is architecturally simple but commercially inconvenient. Most agent platforms store credentials permanently because it's easier to build and easier to support. \"Enter your API key once and forget about it\" is a better user experience than \"your credential expired and needs to be retrieved from the vault.\" The security trade-off is invisible to the user until a breach happens.",[14,827,828],{},"We chose the harder UX because the alternative is indefensible. Microsoft's security blog explicitly warned against running OpenClaw on work machines, partly because of the credential storage model. Kaspersky documented that infostealers are already targeting these files. CrowdStrike's enterprise advisory flagged credential exposure as a primary risk.",[14,830,831],{},"Every AI agent platform will eventually implement some form of credential TTL. The question is whether they do it before or after a major breach forces them to. We chose before.",[14,833,834],{},"The broader lesson extends beyond AI agents. Any system that stores third-party credentials indefinitely is creating a compounding risk that grows every day. The longer the credential sits, the more opportunities an attacker has to reach it. Time-limited credentials aren't a new concept (JWT tokens expire, OAuth refresh tokens rotate, session cookies timeout). AI agents are the last category of software that still stores credentials like it's 2005.",[14,836,591,837,841],{},[183,838,840],{"href":839},"/blog/openclaw-security-checklist","complete security checklist for self-hosted OpenClaw deployments",", our checklist covers manual credential rotation as a partial mitigation for users who can't implement auto-purge.",[14,843,844,845,849],{},"If you want secrets auto-purge, verified skills, and sandboxed execution without building the architecture yourself, ",[183,846,848],{"href":449,"rel":847},[451],"give BetterClaw a try",". Free tier with 1 agent and BYOK. $19/month per agent for Pro. The credentials purge automatically. The skills are pre-vetted. The execution is sandboxed. The security isn't a configuration you maintain. It's a foundation you stand on.",[14,851,852],{},[100,853],{"alt":854,"src":855},"Secrets auto-purge is architecturally simple but commercially inconvenient. Here is why nobody does it.","/img/blog/ai-agent-secrets-auto-purge-tradeoff.jpg",[36,857,467],{"id":466},[14,859,860],{},[52,861,862],{},"What is secrets auto-purge in AI agents?",[14,864,865],{},"Secrets auto-purge is a security architecture where credentials (API keys, OAuth tokens, bot tokens) are automatically erased from an AI agent's memory after a fixed time window, typically 5 minutes. The agent retrieves credentials from an encrypted vault when needed, uses them for the task, and the credentials are purged after the TTL expires. This reduces the credential exposure window from \"forever\" (OpenClaw default) to minutes.",[14,867,868],{},[52,869,870],{},"Why does OpenClaw store API keys in plaintext?",[14,872,873,874,876],{},"OpenClaw stores credentials in ",[56,875,581],{}," as plaintext JSON files. This was a design choice prioritizing simplicity over security. Kaspersky confirmed this in their audit, noting that configuration, memory, and chat logs store API keys and passwords in plain text. RedLine and Lumma infostealers have already added OpenClaw file paths to their must-steal lists. Microsoft's security blog recommended against running OpenClaw on personal or corporate machines partly because of this.",[14,878,879],{},[52,880,881],{},"How does secrets auto-purge protect against credential theft?",[14,883,884],{},"Auto-purge reduces the attack window from permanent to 5 minutes. If an agent uses credentials for 3 tasks per day at 5 minutes each, total daily exposure is 15 minutes versus 1,440 minutes (24 hours) on OpenClaw. A malicious skill or vulnerability that accesses the agent's memory after the purge window finds no credentials. Combined with verified skills and Docker sandboxing, this addresses the full attack chain from access to exfiltration.",[14,886,887],{},[52,888,889],{},"Is 5 minutes enough time for AI agent tasks?",[14,891,892],{},"Yes. 99%+ of single-task workflows (API calls, email searches, message sends, data lookups) complete within 2-3 minutes. The 5-minute TTL provides buffer for multi-step workflows that chain several API calls. Shorter windows (30 seconds) were tested but caused failures in legitimate multi-service workflows. 5 minutes balances security (96% reduction in exposure) with functionality.",[14,894,895],{},[52,896,897],{},"Does BetterClaw encrypt stored credentials?",[14,899,900],{},"Yes. Credentials are stored in an encrypted vault using AES-256 encryption, not in plaintext files. They're retrieved from the vault only when needed for a specific task, provided to the agent in memory, and purged after 5 minutes. Even at rest in the vault, credentials are encrypted. This is layered with Docker-sandboxed execution (skills can't access the vault directly) and verified skills (malicious skills aren't installed in the first place).",{"title":504,"searchDepth":505,"depth":505,"links":902},[903,904,905,906,907,908,909],{"id":571,"depth":505,"text":572},{"id":605,"depth":505,"text":606},{"id":665,"depth":505,"text":666},{"id":690,"depth":505,"text":691},{"id":779,"depth":505,"text":780},{"id":818,"depth":505,"text":819},{"id":466,"depth":505,"text":467},"2026-04-30","OpenClaw stores API keys in plaintext forever. A 5-minute auto-purge reduces exposure by 96%. Here's the attack it prevents and the architecture behind it.","/img/blog/ai-agent-secrets-auto-purge.jpg",{},"7 min read",{"title":553,"description":911},"Secrets Auto-Purge: AI Agent Security for API Keys","blog/ai-agent-secrets-auto-purge",[919,920,921,922,923,924,925,926,927,928],"AI agent security","secrets auto-purge","API key security AI agent","OpenClaw plaintext credentials","credential TTL","AI agent credential exposure","agent memory security","OpenClaw security","BetterClaw security","AES-256 encryption","GRtPlaSa56wZMFFL5azk66a9YNG2pTFTxHAYQsZkW8U",{"id":931,"title":932,"author":933,"body":934,"category":527,"date":1295,"description":1296,"extension":530,"featured":531,"image":1297,"imageHeight":533,"imageWidth":533,"meta":1298,"navigation":535,"path":1299,"readingTime":1300,"seo":1301,"seoTitle":1302,"stem":1303,"tags":1304,"updatedDate":1295,"__hash__":1311},"blog/blog/anthropic-ai-bank-cyber-risk.md","Anthropic's Mythos Just Got Bank CEOs Summoned to Washington. Here's What It Means for Your AI Agents.",{"name":7,"role":8,"avatar":9},{"type":11,"value":935,"toc":1283},[936,942,945,948,951,954,957,961,964,967,970,973,979,983,986,989,992,995,998,1001,1005,1008,1011,1018,1021,1027,1030,1036,1040,1042,1045,1048,1051,1054,1062,1074,1078,1081,1084,1087,1090,1098,1104,1108,1111,1114,1117,1120,1126,1130,1133,1139,1145,1156,1162,1168,1172,1175,1178,1181,1196,1199,1201,1206,1209,1214,1217,1222,1228,1233,1236,1241,1244,1248],[14,937,938],{},[939,940,941],"em",{},"The collision of frontier AI models and financial infrastructure is rewriting the rules of cyber risk. If you're running AI agents, you're already in the blast radius.",[14,943,944],{},"Treasury Secretary Scott Bessent and Fed Chair Jerome Powell pulled bank CEOs into an emergency meeting this week. Not about interest rates. Not about a liquidity crisis.",[14,946,947],{},"About an AI model.",[14,949,950],{},"Anthropic's Claude Mythos, a frontier model so capable at finding software vulnerabilities that the company warned its own government contacts it would make large-scale cyberattacks \"much more likely in 2026.\" The model identified thousands of zero-day vulnerabilities in its first weeks of testing, many of them one to two decades old, hiding in the software that runs everything from hospital networks to trading floors.",[14,952,953],{},"If you're building or deploying AI agents right now, this isn't some abstract policy story. This is the environment your agents are operating in.",[14,955,956],{},"And it's about to get a lot more hostile.",[36,958,960],{"id":959},"the-moment-ai-cyber-risk-stopped-being-theoretical","The moment AI cyber risk stopped being theoretical",[14,962,963],{},"Let's rewind to September 2025. Anthropic detected what analysts now call the first fully autonomous AI espionage campaign at scale. A Chinese state-sponsored group used agentic AI capabilities to conduct vulnerability discovery, lateral movement, and payload execution with minimal human oversight.",[14,965,966],{},"Read that again. Minimal human oversight. An AI agent, not a team of hackers, ran the operation.",[14,968,969],{},"Then in January 2026, a Russian-speaking cybercriminal with limited technical skills used Claude and DeepSeek to hack over 600 devices across 55 countries. According to AWS's security research team, the attacker used generative AI to scale well-known attack techniques throughout every phase of their operation. At one point, the attacker asked Claude in Russian to build a web panel for managing hundreds of targets.",[14,971,972],{},"This is the new baseline. Not nation-state hackers with decades of training. Script kiddies with API keys.",[14,974,975],{},[100,976],{"alt":977,"src":978},"Timeline of AI-powered cyber attacks from September 2025 autonomous espionage to January 2026 mass exploitation","/img/blog/anthropic-ai-bank-cyber-risk-timeline.jpg",[36,980,982],{"id":981},"why-mythos-changes-the-math-for-everyone","Why Mythos changes the math for everyone",[14,984,985],{},"Here's the part that should make you uncomfortable.",[14,987,988],{},"Current AI models can identify high-severity vulnerabilities. Mythos can find five separate vulnerabilities in a single piece of software and chain them together into a novel attack that no human security team would have anticipated. Coupled with the ability to work unsupervised for extended periods, Anthropic says we've hit an inflection point.",[14,990,991],{},"Shlomo Kramer, founder and CEO of Cato Networks, put it bluntly: the agentic attackers are coming and this is a watershed event in the history of cybersecurity. Cisco's chief security officer Anthony Grieco said the old ways of hardening systems are no longer sufficient.",[14,993,994],{},"And here's what nobody tells you: the window is narrow. Alex Stamos, chief product officer at cybersecurity firm Corridor, estimates the open-source models will catch up to frontier model bug-finding capabilities within six months.",[14,996,997],{},"The attackers only need to find one way in. Defenders have to cover every surface.",[14,999,1000],{},"That asymmetry has always existed in cybersecurity. AI just compressed the timeline from months to minutes.",[36,1002,1004],{"id":1003},"what-this-means-if-youre-running-ai-agents","What this means if you're running AI agents",[14,1006,1007],{},"Stay with me here, because this is where it gets personal.",[14,1009,1010],{},"If you're self-hosting an OpenClaw agent on a VPS, a DigitalOcean droplet, or even a Mac Mini under your desk, your attack surface just expanded dramatically. Every exposed port, every unpatched dependency, every misconfigured Docker container is now a target that can be discovered and exploited at machine speed.",[14,1012,1013,1014,1017],{},"The ",[183,1015,1016],{"href":594},"OpenClaw security risks"," we've been writing about for months aren't hypothetical anymore. They're the exact kind of vulnerabilities that Mythos-class models will find and chain together.",[14,1019,1020],{},"Think about what a typical self-hosted agent setup looks like:",[14,1022,1023,1024,1026],{},"Docker containers with default configurations. API keys stored in ",[56,1025,637],{}," files. Ports exposed to the public internet. No intrusion detection. No automated patching. No audit logging.",[14,1028,1029],{},"That was \"good enough\" when the threat was a bored teenager with Metasploit. It is not good enough when the threat is an autonomous AI agent running 24/7 vulnerability scans.",[14,1031,1032],{},[100,1033],{"alt":1034,"src":1035},"Self-hosted AI agent attack surface showing exposed ports, unpatched dependencies, and plaintext credentials","/img/blog/anthropic-ai-bank-cyber-risk-attack-surface.jpg",[36,1037,1039],{"id":1038},"the-infrastructure-gap-most-agent-builders-ignore","The infrastructure gap most agent builders ignore",[14,1041,609],{},[14,1043,1044],{},"They think security is something you bolt on after your agent works. First get the YAML right. First get the skills installed. First get the model routing figured out. Security can wait.",[14,1046,1047],{},"It can't wait anymore.",[14,1049,1050],{},"Anthropic launched Project Glasswing alongside Mythos, giving 12 partner organizations including Microsoft, Apple, and Cisco early access to find and fix vulnerabilities before they get exploited. That tells you something about the urgency.",[14,1052,1053],{},"But most teams running AI agents aren't Microsoft. They don't have a dedicated security team scanning their infrastructure. They're a founder, a small dev team, maybe a contractor. They're choosing between building features and patching CVEs.",[14,1055,1056,1057,1061],{},"If you've been wrestling with ",[183,1058,1060],{"href":1059},"/blog/openclaw-docker-troubleshooting","OpenClaw Docker troubleshooting"," or spending weekends maintaining your agent infrastructure, this is the moment to ask yourself: is that really how you want to spend your time in a world where AI-powered attacks operate at machine speed?",[14,1063,1064,1065,1069,1070,1073],{},"We built ",[183,1066,1068],{"href":1067},"/","Better Claw"," because we were tired of infrastructure eating our weekends. But in light of what Anthropic just disclosed, managed hosting isn't just about convenience anymore. It's about not being the low-hanging fruit in an environment where autonomous attackers are scanning for exactly that. ",[183,1071,1072],{"href":461},"$19/month per agent",", and your infrastructure is somebody else's problem.",[36,1075,1077],{"id":1076},"what-the-bessent-powell-meeting-actually-signals","What the Bessent-Powell meeting actually signals",[14,1079,1080],{},"And that's when we realized this story isn't really about banks.",[14,1082,1083],{},"Yes, Bessent and Powell summoned Wall Street CEOs to make sure financial institutions are preparing defenses against Mythos-class threats. But the real signal is simpler: the US government now considers AI-generated cyber risk a systemic threat.",[14,1085,1086],{},"Not a \"keep an eye on it\" threat. A \"clear your calendar and come to Washington\" threat.",[14,1088,1089],{},"The implications cascade downward. If banks need to harden their systems, every vendor and partner in their supply chain needs to do the same. If you're building an AI agent that touches financial data, customer PII, or payment systems, the security bar just jumped by an order of magnitude.",[14,1091,1092,1093,1097],{},"This is especially relevant if you're running agents for ",[183,1094,1096],{"href":1095},"/blog/openclaw-agents-for-ecommerce","ecommerce use cases"," or anything that handles customer data. The regulatory scrutiny that follows a story like this always trickles down.",[14,1099,1100],{},[100,1101],{"alt":1102,"src":1103},"Cascade of AI cyber risk regulations from government to banks to vendors to AI agent builders","/img/blog/anthropic-ai-bank-cyber-risk-cascade.jpg",[36,1105,1107],{"id":1106},"the-arms-race-youre-already-part-of","The arms race you're already part of",[14,1109,1110],{},"But that's not even the real problem.",[14,1112,1113],{},"Every major AI lab's next model will push cyber capabilities further. Behind Mythos is the next OpenAI model, and the next Gemini, and a few months behind them are the open-source Chinese models. As Kramer told CNN, the defenders need to run as fast as they can just to stay in the same place.",[14,1115,1116],{},"This creates a permanent tax on every team running AI infrastructure. You need automated patching. You need encrypted secrets management. You need isolated execution environments. You need audit logs. You need somebody watching the monitors at 3 AM when a Mythos-inspired scanner finds a forgotten port.",[14,1118,1119],{},"Or you need to outsource that entire burden.",[14,1121,1013,1122,1125],{},[183,1123,1124],{"href":839},"OpenClaw security checklist"," we published is a good starting point if you're committed to self-hosting. But be honest with yourself about whether you can maintain that posture indefinitely against adversaries that don't sleep, don't get bored, and don't make typos.",[36,1127,1129],{"id":1128},"what-to-actually-do-right-now","What to actually do right now",[14,1131,1132],{},"Let me be practical. Here's what matters this week, not this quarter.",[14,1134,1135,1138],{},[52,1136,1137],{},"Audit your exposed surfaces."," If your agent is reachable from the public internet, assume it will be scanned by something smarter than you within days. Check every open port. Check your Docker configs. Check where your API keys live.",[14,1140,1141,1144],{},[52,1142,1143],{},"Update everything."," Mythos found vulnerabilities that were one to two decades old. The boring stuff matters more than ever.",[14,1146,1147,1150,1151,1155],{},[52,1148,1149],{},"Evaluate your hosting model."," Self-hosting made sense when the primary risk was downtime. The risk profile has changed. Consider whether ",[183,1152,1154],{"href":1153},"/openclaw-hosting","managed OpenClaw hosting"," is worth the tradeoff.",[14,1157,1158,1161],{},[52,1159,1160],{},"Watch the regulatory signals."," The Bessent-Powell meeting is the first domino. If you're building agents for regulated industries, expect compliance requirements to tighten fast.",[14,1163,1164,1167],{},[52,1165,1166],{},"Don't panic, but don't ignore this."," The fact that Anthropic launched Project Glasswing means the industry is taking this seriously. The worst response is to assume you're too small to be a target. Automated attacks don't discriminate by company size.",[36,1169,1171],{"id":1170},"the-honest-takeaway","The honest takeaway",[14,1173,1174],{},"Here's what I keep coming back to.",[14,1176,1177],{},"We got into AI agents because the technology is genuinely exciting. Watching an agent autonomously handle tasks that used to take hours of manual work is one of the best feelings in tech right now. That hasn't changed.",[14,1179,1180],{},"What's changed is the environment. The same agentic capabilities that make our tools powerful also make the threats against our infrastructure more capable. That's not a reason to stop building. It's a reason to build on foundations that can withstand what's coming.",[14,1182,1183,1184,1186,1187,1191,1192,1195],{},"If any of this hit close to home, if you've been running a self-hosted agent and putting off the security hardening, if you know your ",[56,1185,637],{}," file is doing more heavy lifting than it should, ",[183,1188,1190],{"href":449,"rel":1189},[451],"give Better Claw a look",". It's $19/month per agent, BYOK, and you get managed infrastructure with security that doesn't depend on you remembering to run ",[56,1193,1194],{},"apt update"," at midnight. We handle the infrastructure. You handle the interesting part.",[14,1197,1198],{},"The agentic attackers are coming. Make sure your agents are ready.",[36,1200,467],{"id":466},[14,1202,1203],{},[52,1204,1205],{},"What is the Anthropic Mythos AI model and why does it matter for cyber risk?",[14,1207,1208],{},"Claude Mythos is Anthropic's most powerful AI model to date, sitting above its Opus tier. It matters because it can autonomously discover, chain together, and exploit software vulnerabilities at speeds no human team can match. In its first weeks of testing, it found thousands of zero-day flaws, many hidden for over a decade.",[14,1210,1211],{},[52,1212,1213],{},"How does AI-driven cyber risk affect banks and financial services?",[14,1215,1216],{},"Treasury Secretary Bessent and Fed Chair Powell summoned bank CEOs specifically over Mythos-class threats, signaling the government views AI cyber risk as systemic to financial stability. Banks face pressure to harden systems across their entire supply chain, which cascades to every vendor and partner handling financial data.",[14,1218,1219],{},[52,1220,1221],{},"How do I secure my self-hosted AI agent against AI-powered attacks?",[14,1223,1224,1225,1227],{},"Start by auditing exposed ports, moving secrets out of ",[56,1226,637],{}," files into encrypted vaults, keeping all dependencies patched, and enabling audit logging. If maintaining that security posture continuously isn't realistic for your team, evaluate managed hosting options that handle infrastructure security for you.",[14,1229,1230],{},[52,1231,1232],{},"Is managed AI agent hosting worth the cost for security alone?",[14,1234,1235],{},"At $19/month per agent, managed hosting like BetterClaw costs less than a single hour of incident response consulting. You get isolated environments, automated updates, encrypted secrets management, and monitoring without needing to maintain it yourself. In a world of autonomous AI-powered scanning, the cost of a breach far exceeds the cost of prevention.",[14,1237,1238],{},[52,1239,1240],{},"Is my small project really a target for AI-powered cyberattacks?",[14,1242,1243],{},"Yes. Automated scanning tools, including the techniques Mythos enables, don't discriminate by company size. In January 2026, a single attacker with limited skills used AI to compromise 600+ devices across 55 countries. If your agent is reachable from the internet, it's a target regardless of how small your operation is.",[36,1245,1247],{"id":1246},"related-reading","Related Reading",[199,1249,1250,1256,1262,1269,1276],{},[202,1251,1252,1255],{},[183,1253,1254],{"href":594},"OpenClaw Security Risks Explained"," — The specific vulnerabilities AI attackers will target",[202,1257,1258,1261],{},[183,1259,1260],{"href":839},"OpenClaw Security Checklist"," — Hardening steps if you're committed to self-hosting",[202,1263,1264,1268],{},[183,1265,1267],{"href":1266},"/blog/openclaw-gateway-guide","OpenClaw Gateway Guide"," — The single setting that exposed 30,000+ instances",[202,1270,1271,1275],{},[183,1272,1274],{"href":1273},"/blog/openclaw-skill-audit","OpenClaw Skill Audit"," — How to check for compromised skills in your setup",[202,1277,1278,1282],{},[183,1279,1281],{"href":1280},"/compare/openclaw","BetterClaw vs Self-Hosted OpenClaw"," — Managed security vs DIY in the new threat landscape",{"title":504,"searchDepth":505,"depth":505,"links":1284},[1285,1286,1287,1288,1289,1290,1291,1292,1293,1294],{"id":959,"depth":505,"text":960},{"id":981,"depth":505,"text":982},{"id":1003,"depth":505,"text":1004},{"id":1038,"depth":505,"text":1039},{"id":1076,"depth":505,"text":1077},{"id":1106,"depth":505,"text":1107},{"id":1128,"depth":505,"text":1129},{"id":1170,"depth":505,"text":1171},{"id":466,"depth":505,"text":467},{"id":1246,"depth":505,"text":1247},"2026-04-10","Anthropic's Mythos model triggered an emergency bank CEO meeting. Learn what AI-driven cyber risk means for your AI agents and how to protect them.","/img/blog/anthropic-ai-bank-cyber-risk.jpg",{},"/blog/anthropic-ai-bank-cyber-risk","10 min read",{"title":932,"description":1296},"Anthropic AI Cyber Risk: What Bank CEO Warnings Mean for Agents","blog/anthropic-ai-bank-cyber-risk",[1305,1306,1307,1308,1309,1310],"anthropic ai cyber risk","mythos ai model security","ai agent security","openclaw security","ai cybersecurity threats","managed ai agent hosting","abtd9SFcnUzwrrV244DKKIdQ2617mNqBFS6kn58IlZc",{"id":1313,"title":1314,"author":1315,"body":1316,"category":527,"date":1682,"description":1683,"extension":530,"featured":531,"image":1684,"imageHeight":533,"imageWidth":533,"meta":1685,"navigation":535,"path":1686,"readingTime":537,"seo":1687,"seoTitle":1688,"stem":1689,"tags":1690,"updatedDate":1682,"__hash__":1698},"blog/blog/best-openclaw-skills-2026.md","Best OpenClaw Skills to Install in 2026 (and 5 That Will Steal Your API Keys)",{"name":7,"role":8,"avatar":9},{"type":11,"value":1317,"toc":1664},[1318,1321,1324,1327,1330,1333,1336,1340,1346,1350,1356,1362,1368,1372,1383,1389,1395,1399,1405,1416,1422,1428,1434,1438,1441,1447,1453,1463,1469,1475,1481,1485,1491,1497,1503,1513,1519,1529,1532,1536,1542,1545,1551,1557,1563,1569,1575,1578,1586,1590,1593,1596,1599,1602,1605,1612,1614,1618,1621,1625,1628,1632,1650,1654,1657,1661],[14,1319,1320],{},"ClawHub has 10,700+ skills. 820+ are flagged malicious. A Snyk audit found 13.4% have critical security issues. Here are the 10 worth installing, the 5 red flags to watch for, and why \"just browse ClawHub\" is dangerous advice.",[14,1322,1323],{},"Cisco's security team found an OpenClaw skill performing data exfiltration without the user's knowledge. Not a theoretical vulnerability. Not a proof-of-concept. A skill on ClawHub, with real installs, quietly sending user data to an external server.",[14,1325,1326],{},"The user had no idea.",[14,1328,1329],{},"That's the ClawHub problem in one sentence. 10,700+ community-built skills. No mandatory code review. No security testing before publication. You search for \"Gmail automation,\" install something that looks right, and hope nobody slipped a credential harvester into the code.",[14,1331,1332],{},"In January 2026, researchers discovered ClawHavoc: 341 malicious skills using typosquatted names (\"clawhubb\" instead of \"clawhub\") distributing Atomic Stealer malware that exfiltrated SSH keys, API tokens, and browser cookies. ClawHub removed 2,419 suspicious skills and partnered with VirusTotal for scanning. But \"partnered with VirusTotal\" is not \"every skill is verified.\"",[14,1334,1335],{},"Here are the 10 skills actually worth installing, the 5 red flags that indicate a skill is malicious, and the verification checklist you should run before trusting anything on ClawHub.",[36,1337,1339],{"id":1338},"the-10-openclaw-skills-worth-installing-verified-safe-actively-maintained","The 10 OpenClaw skills worth installing (verified safe, actively maintained)",[14,1341,1342],{},[100,1343],{"alt":1344,"src":1345},"Three-tier ranking of OpenClaw skills worth installing: foundation, productivity, and power user","/img/blog/best-openclaw-skills-2026-top-ten.jpg",[143,1347,1349],{"id":1348},"tier-1-install-these-first-everyone-needs-them","Tier 1: Install these first (everyone needs them)",[14,1351,1352,1355],{},[52,1353,1354],{},"1. Web Browsing (official, 180K+ installs)."," The most-installed skill on ClawHub. Your agent can navigate pages, extract content, and follow links. Without this, the agent is a chatbot running on stale training data. First-party skill. Zero registry risk.",[14,1357,1358,1361],{},[52,1359,1360],{},"2. Web Search (Tavily or Brave)."," Web Browsing fetches pages. Web Search finds them. Tavily returns LLM-friendly structured results (1,000 free searches/month). Brave uses an independent index with freshness filtering ($5/month free credit). Pick one. Install both if you need coverage.",[14,1363,1364,1367],{},[52,1365,1366],{},"3. Telegram Gateway."," The second most-installed skill (145K+ installs). Connect with a BotFather token. Message your agent from your phone. Setup: 5 minutes. Latency: surprisingly low. This is how 80% of OpenClaw users interact with their agent daily.",[143,1369,1371],{"id":1370},"tier-2-productivity-install-after-your-agent-is-stable","Tier 2: Productivity (install after your agent is stable)",[14,1373,1374,1377,1378,1382],{},[52,1375,1376],{},"4. GOG (Google Workspace, 14K+ installs)."," Gmail, Calendar, Drive, Docs, Sheets, Contacts through a single OAuth connection. For the complete Gmail and Calendar automation guide, our ",[183,1379,1381],{"href":1380},"/blog/openclaw-gmail-calendar-automation","guide"," covers the five workflows that save the most time.",[14,1384,1385,1388],{},[52,1386,1387],{},"5. Screenshot & OCR."," Captures screenshots and extracts text using optical character recognition. 4/5 stars on ClawHub. Useful for documentation workflows, reading error dialogs, and digitizing printed text. Runs locally. No cloud API calls. Supports 60+ languages.",[14,1390,1391,1394],{},[52,1392,1393],{},"6. Task Automation (cron)."," Schedule recurring tasks. Morning briefings at 7 AM. Weekly reports on Monday. Daily inbox triage. The backbone of any always-on agent workflow.",[143,1396,1398],{"id":1397},"tier-3-power-user-install-only-if-you-need-the-specific-capability","Tier 3: Power user (install only if you need the specific capability)",[14,1400,1401,1404],{},[52,1402,1403],{},"7. N8N Workflow Automation."," Connects OpenClaw to your N8N instance. Trigger complex multi-step workflows from chat. Runs locally. Data stays private. The bridge between your agent and your automation stack.",[14,1406,1407,1410,1411,1415],{},[52,1408,1409],{},"8. ElevenLabs Voice Agent."," Gives OpenClaw a voice. The fail-safe mechanism is clever: if text/email fails, the bot automatically calls the recipient. For the OpenClaw voice agent setup with Twilio, our ",[183,1412,1414],{"href":1413},"/blog/openclaw-voice-agents-twilio-meet","comparison"," covers how voice integrates with other channels.",[14,1417,1418,1421],{},[52,1419,1420],{},"9. Capability Evolver (35K+ installs)."," The agent automatically improves its own capabilities during operation. Top of the ClawHub charts. Useful for long-running agents that need to adapt over time.",[14,1423,1424,1427],{},[52,1425,1426],{},"10. Home Assistant."," Controls your smart home through natural language. No cloud dependency. No data leaving your network. Privacy-first home automation. \"Which devices are currently on?\" works as a chat command.",[14,1429,1430,1433],{},[52,1431,1432],{},"The bundled skills rule:"," 53 skills ship bundled with OpenClaw as first-party plugins. These carry zero registry risk. Start with bundled skills. Add ClawHub skills only when bundled options don't cover your use case.",[36,1435,1437],{"id":1436},"the-5-red-flags-that-mean-a-skill-is-malicious","The 5 red flags that mean a skill is malicious",[14,1439,1440],{},"Here's what nobody tells you about ClawHub security.",[14,1442,1443],{},[100,1444],{"alt":1445,"src":1446},"Five red flags for malicious OpenClaw skills: no GitHub link, wildcard permissions, typosquatted name, stale and low installs, missing VirusTotal scan","/img/blog/best-openclaw-skills-2026-red-flags.jpg",[14,1448,1449,1452],{},[52,1450,1451],{},"Red Flag 1: No linked GitHub repository."," Legitimate skills link to a maintained GitHub repo with issues, PRs, and commit history. If the ClawHub listing has no source code link, the author doesn't want you reading the code. That's the first sign.",[14,1454,1455,1458,1459,1462],{},[52,1456,1457],{},"Red Flag 2: Wildcard shell permissions on a simple skill."," A weather skill that requests ",[56,1460,1461],{},"Bash(*)"," access can execute any command on your system. The permission model exists. Use it. Read what the skill requests before confirming the install.",[14,1464,1465,1468],{},[52,1466,1467],{},"Red Flag 3: Name mimics a popular skill (typosquatting)."," ClawHavoc used names like \"clawhubb\" instead of \"clawhub.\" Check the spelling carefully. Verify the publisher. If the name is one character off from a popular skill, it's almost certainly malicious.",[14,1470,1471,1474],{},[52,1472,1473],{},"Red Flag 4: Stale with low installs."," Last updated 3+ months ago. Fewer than 100 installs. No reviews. This is either abandoned or was published as a one-time attack payload. Both are reasons not to install.",[14,1476,1477,1480],{},[52,1478,1479],{},"Red Flag 5: Missing or non-benign VirusTotal scan."," Since February 2026, ClawHub shows VirusTotal scan results on each skill's page. If the scan says anything other than \"Benign,\" don't install. If there's no scan result at all, the skill predates the VirusTotal partnership and hasn't been re-scanned.",[36,1482,1484],{"id":1483},"the-verification-checklist-run-this-before-every-clawhub-install","The verification checklist (run this before every ClawHub install)",[14,1486,1487],{},[100,1488],{"alt":1489,"src":1490},"Five-step ClawHub skill verification checklist: VirusTotal scan, GitHub repo, SKILL.md permissions, install count and reviews, sandbox mode test","/img/blog/best-openclaw-skills-2026-verification-checklist.jpg",[14,1492,1493,1496],{},[52,1494,1495],{},"Step 1:"," Check the VirusTotal scan on the ClawHub page. Must show \"Benign.\"",[14,1498,1499,1502],{},[52,1500,1501],{},"Step 2:"," Click the GitHub repository link. Verify real commits, real issues, a real maintainer with other projects.",[14,1504,1505,1508,1509,1512],{},[52,1506,1507],{},"Step 3:"," Read the ",[56,1510,1511],{},"SKILL.md",". What permissions does it request? A calendar skill doesn't need shell access. A search skill doesn't need file write.",[14,1514,1515,1518],{},[52,1516,1517],{},"Step 4:"," Check the install count and reviews. Under 100 installs with zero reviews is a warning sign.",[14,1520,1521,1524,1525,1528],{},[52,1522,1523],{},"Step 5:"," Enable sandbox mode before testing: ",[56,1526,1527],{},"openclaw config --global --sandbox=strict",". Test the skill with non-sensitive data first.",[14,1530,1531],{},"If running a 5-step security verification on every skill you install, monitoring ClawHub for typosquatted malware, reading SKILL.md permission manifests, and manually sandboxing untrusted code sounds like more security work than productivity work, BetterClaw's verified skills marketplace handles all of this. Every skill tested by our team before publication. No ClawHub registry risk. No VirusTotal manual checks. No malicious skill surprises. Free tier with 1 agent and BYOK. $19/month per agent for Pro.",[36,1533,1535],{"id":1534},"the-security-numbers-that-should-concern-you","The security numbers that should concern you",[14,1537,1538],{},[100,1539],{"alt":1540,"src":1541},"ClawHub security statistics: 10,700+ skills, 820+ flagged malicious, 13.4% with critical issues, 341 ClawHavoc payloads, 1 in 5 malicious pre-cleanup","/img/blog/best-openclaw-skills-2026-security-stats.jpg",[14,1543,1544],{},"Here's the data.",[14,1546,1547,1550],{},[52,1548,1549],{},"10,700+ skills"," on ClawHub as of May 2026. After the ClawHavoc cleanup removed 2,419, the count has grown back.",[14,1552,1553,1556],{},[52,1554,1555],{},"820+ flagged malicious"," by Cisco's analysis. That's 7.6% of the registry.",[14,1558,1559,1562],{},[52,1560,1561],{},"13.4% have critical issues"," according to Snyk's independent audit. Malware, prompt injection, exposed API keys.",[14,1564,1565,1568],{},[52,1566,1567],{},"341 skills distributed Atomic Stealer malware"," in the ClawHavoc campaign. Typosquatted names. Reverse shells. SSH key exfiltration.",[14,1570,1571,1574],{},[52,1572,1573],{},"1 in 5 skills were malicious"," before the February 2026 cleanup, according to Koi Security's audit of 2,857 skills.",[14,1576,1577],{},"The math: If you install 10 random skills from ClawHub without vetting, statistically 1-2 of them have critical security issues. That's not a theoretical risk. That's a probability.",[14,1579,1580,1581,1585],{},"For the complete OpenClaw security analysis including CVEs and enterprise risks, our ",[183,1582,1584],{"href":1583},"/blog/openclaw-security-2026","security guide"," covers the broader attack surface.",[36,1587,1589],{"id":1588},"the-honest-take-why-we-built-a-verified-marketplace-instead","The honest take (why we built a verified marketplace instead)",[14,1591,1592],{},"Here's the perspective.",[14,1594,1595],{},"ClawHub is the most feature-rich skill ecosystem in AI agents. 10,700+ skills covering every use case imaginable. But an ecosystem with 7.6% malicious content and no mandatory code review is fundamentally different from an ecosystem where every skill is tested before publication.",[14,1597,1598],{},"We built BetterClaw's verified marketplace specifically because of ClawHavoc. After seeing 341 malicious skills steal SSH keys from real users, we decided that \"user beware\" is not an acceptable security model for an agent that has access to your email, calendar, files, and messaging apps.",[14,1600,1601],{},"The trade-off is real. BetterClaw's verified marketplace has fewer skills than ClawHub. We test every one. That takes time. The breadth is smaller. But every skill in our marketplace is verified safe. No credential harvesters. No cryptominers. No prompt injection. No typosquatting.",[14,1603,1604],{},"The 10 skills listed above are genuinely excellent. Install them with confidence after running the verification checklist. For everything else on ClawHub, proceed with the same caution you'd apply to installing a random npm package from an anonymous publisher on day one.",[14,1606,1607,1608,1611],{},"If you want the skill ecosystem without the security lottery, ",[183,1609,848],{"href":449,"rel":1610},[451],". Free tier with 1 agent and BYOK. $19/month per agent for Pro. Verified skills marketplace. Every skill tested. Zero ClawHub registry risk. The agent does the work. The skills are safe.",[36,1613,467],{"id":466},[143,1615,1617],{"id":1616},"what-are-the-best-openclaw-skills-to-install-in-2026","What are the best OpenClaw skills to install in 2026?",[14,1619,1620],{},"The top 10 by safety and utility: Web Browsing (180K installs, first-party), Web Search (Tavily or Brave), Telegram Gateway (145K installs), GOG Google Workspace (14K installs), Screenshot & OCR, Task Automation (cron), N8N Workflow, ElevenLabs Voice, Capability Evolver (35K installs), and Home Assistant. Start with the 53 bundled first-party skills before adding ClawHub community skills.",[143,1622,1624],{"id":1623},"are-openclaw-skills-on-clawhub-safe","Are OpenClaw skills on ClawHub safe?",[14,1626,1627],{},"Not all of them. 820+ skills (7.6%) are flagged malicious. A Snyk audit found 13.4% have critical security issues. The ClawHavoc campaign distributed 341 malicious skills with Atomic Stealer malware through typosquatted names. ClawHub partnered with VirusTotal for scanning in February 2026, but verification is not mandatory. Always check the VirusTotal status, GitHub repo, permissions, and install count before installing.",[143,1629,1631],{"id":1630},"how-do-i-install-openclaw-skills-from-clawhub","How do I install OpenClaw skills from ClawHub?",[14,1633,1634,1635,1638,1639,1642,1643,1646,1647,1649],{},"Run ",[56,1636,1637],{},"npx clawhub@latest install \u003Cskill-name>"," from your terminal. The installer shows the permissions the skill requests. Review them before confirming. After installation, run ",[56,1640,1641],{},"npx clawhub@latest list"," to see installed skills and ",[56,1644,1645],{},"npx clawhub@latest update"," to update all. Enable sandbox mode (",[56,1648,1527],{},") before testing untrusted skills.",[143,1651,1653],{"id":1652},"how-much-do-openclaw-skills-cost","How much do OpenClaw skills cost?",[14,1655,1656],{},"Most ClawHub skills are free. The underlying services may have costs: Tavily search (1,000 free/month, then paid), ElevenLabs voice ($5+/month), N8N (self-hosted free, cloud $24+/month). BetterClaw's verified marketplace is included in all plans (free tier and $19/month Pro). No separate skill costs. No ClawHub registry risk.",[143,1658,1660],{"id":1659},"what-is-the-clawhavoc-attack-and-should-i-be-worried","What is the ClawHavoc attack and should I be worried?",[14,1662,1663],{},"ClawHavoc was a coordinated attack in January 2026 where 341 malicious skills were uploaded to ClawHub using typosquatted names. These skills installed Atomic Stealer malware that exfiltrated SSH keys, API tokens, and browser session cookies via reverse shells. ClawHub removed 2,419 suspicious skills and implemented VirusTotal scanning. You should be cautious, not panicked. Use the verification checklist before every install. Or use BetterClaw's verified marketplace where every skill is tested before publication.",{"title":504,"searchDepth":505,"depth":505,"links":1665},[1666,1671,1672,1673,1674,1675],{"id":1338,"depth":505,"text":1339,"children":1667},[1668,1669,1670],{"id":1348,"depth":512,"text":1349},{"id":1370,"depth":512,"text":1371},{"id":1397,"depth":512,"text":1398},{"id":1436,"depth":505,"text":1437},{"id":1483,"depth":505,"text":1484},{"id":1534,"depth":505,"text":1535},{"id":1588,"depth":505,"text":1589},{"id":466,"depth":505,"text":467,"children":1676},[1677,1678,1679,1680,1681],{"id":1616,"depth":512,"text":1617},{"id":1623,"depth":512,"text":1624},{"id":1630,"depth":512,"text":1631},{"id":1652,"depth":512,"text":1653},{"id":1659,"depth":512,"text":1660},"2026-05-19","10,700+ skills on ClawHub. 820+ are malicious. Here are the 10 worth installing, 5 red flags for dangerous skills, and the verification checklist to stay safe.","/img/blog/best-openclaw-skills-2026.jpg",{},"/blog/best-openclaw-skills-2026",{"title":1314,"description":1683},"Best OpenClaw Skills 2026 (+ 5 to Avoid on ClawHub)","blog/best-openclaw-skills-2026",[1691,1692,1693,1694,1695,1696,1697],"best OpenClaw skills","OpenClaw skills install","ClawHub skills safe","OpenClaw ClawHub malicious","OpenClaw skills 2026","ClawHavoc","OpenClaw skill security","HWDTWyxlkHHn6FFtHVb6l6tlT6fbn5bdtjiQEauyEj4",1780640472543]