Free tools Windows power users keep installed
One-click scans. No signup required.
Stop the agent first. If the change may have been unauthorized—not merely a mistake—restrict the identity or credential it used, preserve logs, and investigate the activity before restoring anything. Then use the affected app’s own undo, version history, backup, or rollback process. There is no universal undo for AI agents: the right response depends on what changed and which account, session, key, or service identity the agent used.
First, stop further actions
Pause or stop the task using the agent’s available control as soon as you see suspicious or harmful behavior. Avoid starting another agent run to “fix” the first one until you know what it can access; that could create more changes. If the platform offers a task or run identifier, note it along with the time you stopped the agent.
As an Amazon Associate I earn from qualifying purchases.
For ChatGPT agent, OpenAI recommends enabling only the apps needed for the current task and regularly reviewing app permissions. See ChatGPT agent.
Recommended Free Tools
Decide whether this is an error or a security incident
An incorrect change can be an ordinary execution error or a misunderstanding of the request. An unauthorized change is different: it may mean the agent was given too much access, someone else used its session or credential, or malicious content influenced its actions. Do not assume every mistake means an account was compromised, but treat unexplained or unauthorized activity as a potential security incident until you have scoped it.
#1 Best Overall
- Built for Local AI Development: AMD Ryzen AI Halo is designed for local AI development and inference, featuring 128GB unified memory and support for up to 200B parameter models to build and run intensive AI workloads locally.
- 128GB Unified Memory: Features 128GB LPDDR5x unified memory at 8000 MT/s with 256 GB/s memory bandwidth, providing a shared memory pool across the CPU, GPU, and NPU to support larger AI models.
- AMD Ryzen AI Max+ 395 Processor: Features 16 cores, 32 threads, and Zen 5 architecture, paired with AMD Radeon 8060S integrated graphics featuring 40 RDNA 3.5 compute units and an AMD XDNA 2 NPU with up to 50 TOPS.
- Linux AI Developer Platform: Purpose-built for Linux-based AI development with full AMD ROCm software support and preloaded tools, models, and workflows optimized for local AI development.
- Compact, Connected Design: Includes a 2TB M.2 SSD, 10GbE LAN, Wi-Fi 7, Bluetooth 5.4, USB-C connectivity, and HDMI 2.1b.
Agent-only systems can act without waiting for a person to approve each action. Google Cloud’s MCP security guidance identifies prompt injection, insecure tool chaining, and weak error handling as risks in agent-only operation; an approval step can reduce exposure but is not foolproof because a person can approve the wrong action. OpenAI’s ChatGPT agent guidance describes malicious content attempting to induce an agent to retrieve and send a password-reset code. See Google Cloud: AI security and safety | Google Cloud MCP servers and OpenAI’s ChatGPT agent guidance.
Contain the access the agent used
Identify the agent’s identity and the access path involved: a connected app permission, logged-in session, API key, service account, or other cloud credential. If unauthorized activity is plausible, restrict or revoke that specific access rather than assuming that stopping the visible run is enough. Use the provider’s current procedure; disabling one identity may not immediately invalidate every token created from it.
If an OpenAI account or API key may be involved
For a suspected ChatGPT account compromise, OpenAI’s Help Center recommends changing an exposed password and logging out all sessions. It notes that logging out other ChatGPT sessions may take up to 30 minutes, and enabling MFA alone does not cancel existing logins. If an API key is implicated, delete the affected key, review API usage, and contact support. Follow the current steps in How can I keep my OpenAI accounts secure?
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Rank #2
- EVOLUTION AMD RYZEN AI MAX+ 395 MINI PC - GMKtec EVO-X2 is the next evolution in AI mini PC Ryzen Strix Halo series. Thanks to AMD Simultaneous Multithreading (SMT) the core-count is effectively doubled, to 32 threads. Ryzen AI Max+ 395 has 64 MB of L3 cache and can boost up to 5.1 GHz, depending on the workload. The Ryzen AI Max+ 395 is currently rated as the "most powerful x86 APU" on the market for AI computing.
- AI NPU with XDNA 2 ARCHITECTURE - Powered by 16 “Zen 5” CPU cores, 50+ peak AI TOPS XDNA 2 NPU and a truly massive integrated GPU driven by 40 AMD RDNA 3.5 CUs, the Ryzen AI MAX+ 395 is a transformative upgrade and delivers a significant performance boost over the competition. The Ryzen AI Max+ 395 excels in consumer AI workloads like the llama.cpp-powered application: LM Studio. Shaping up to be the must-have app for client LLM workloads, LM Studio allows users to locally run the latest language model without any technical knowledge required and unleash their creativity and productivity.
- AMD RADEON 8090S iGPU GAMING PC - The AMD Radeon RX 8060S offers all 40 CUs with up to 2.9 GHz graphics clock and uses the new RDNA 3.5 architecture. The powerful iGPU is positioned between an RTX 4060 and 4070 laptop GPU and therefore enables gaming in FHD at maximum details in most demanding games. The 8060S can also utilize the full 64GB pool, which is perfect for running LLMs such as Deepseek 32B, which runs comfortably on this machine.
- EIGHT CHANNEL LPDDR5X - LPDDR5X is a new ground breaking memory small form factor installed on-board. With blazing speeds up to to 8000MT/s, it runs 1.5x faster than the DDR5 SODIMMs; 90% better performance over DDR5 SODIMMs in video conferencing and photo editing; 30% better performance in productivity apps; 4% better performance in digital content workloads.
- QUAD SCREEN 8K DISPLAY SUPPORT - EVO-X2 AI Mini PC support 4-screen 4K/8K output via HDMI 2.1 (8K@60Hz), DisplayPort 1.4 (4K@60Hz), and dual USB 4 40Gbps Transfer speed (supporting PD3.0/DP1.4/DATA). Ideal for gaming, video editing, and multitasking, it provides expansive and crisp multi-display support.
If a Google Cloud service account key may be involved
Google Cloud’s service-account incident guidance describes replacing a compromised key by deploying a new key, verifying it, disabling the old key, and then deleting it. Google also warns that tokens created from a service account can remain valid for a period after the account is disabled. Its guidance says to wait at least 60 minutes after disabling before re-enabling, or longer if an extended token-lifetime policy applies. That timing is specific to the Google Cloud guidance; do not apply it to other providers or credential types. See Google Cloud: Responding to compromised Google Cloud credentials.
Preserve evidence and work out what changed
Before deleting logs or making broad cleanup changes, save relevant records. Capture the incident time window, agent or run identifier, identity used, affected objects, observed changes, relevant prompts or tool activity, and unusual account or API activity. Keep copies of logs in a place the agent cannot alter. Google Cloud’s AI threat response guidance recommends collecting and backing up logs for affected resources; it also cautions that its informal suggestions may not fit every finding or operation. See Google Cloud: AI threat response.
Next, review the agent’s activity and the resources it could reach during the incident window. Check security history, API usage, application audit logs, and connected resources. For a suspected Google Cloud identity compromise, Google recommends searching for API calls made by that identity and checking for newly created service-account keys, users, or project-level SSH keys. Use the provider’s specific guidance at Responding to compromised Google Cloud credentials.
Rank #3
- Intel Core Ultra 9 285 Processor: Newly developed cores deliver ultra-smooth and responsive gameplay. AI accelerators prepare users for the next era of gaming on an AI PC.
- Simplistic Design: Enjoy the latest generation of Windows 11 Home for your everyday needs. *MSI recommends Windows 11 Pro for business use.
- NVIDIA GeForce RTX 5070 Ti GPU
- Cool While Gaming: In conjunction with an RGB CPU Air Cooler, the Aegis RS features four system cooling fans; three in the front and one in the rear to pull in cool air and push heat out of the PC.
- Turn on the Bright Lights: With the built-in RGB lighting, take your gaming experience to the next level by pressing the MSI LED button to cycle through lighting options. Customize lighting even further with MSI Center software.
- Record what the agent was asked to do and what it actually did.
- List each changed object and any follow-on changes, such as new users, keys, permissions, or messages sent.
- Check whether the same identity or connected app was used for other actions in the same period.
- Keep evidence before cleanup, especially if a provider or organizational security team may need to investigate.
Choose a recovery method for the affected system
After preserving evidence and identifying the affected state, restore from a known-good backup, version history, transaction log, or the application’s undo or rollback feature. The right option depends on the system; the official guidance does not establish one universal procedure for reversing agent changes. Google Cloud’s Office of the CISO recommends rollback infrastructure that can halt multi-agent operations when unexpected behavior is detected. See Google Cloud: Secure AI agents with Google Cloud.
Before restoring, consider how much the agent changed and whether other people or services made legitimate updates afterward. A broad restore may remove valid work; a narrow undo may leave related changes untouched. Compare the practical trade-offs:
| Recovery route | Best suited to | What to check |
|---|---|---|
| Application undo or version history | A discrete edit in an app that retains prior versions or supports undo. | Whether the history includes all affected objects and whether the change can be reversed without overwriting later valid work. |
| Transaction log or targeted rollback | A system that records operations and can reverse selected changes. | Whether related or follow-on actions are included, and whether the log can restore the intended state narrowly. |
| Known-good backup restore | Broader damage where a verified pre-incident copy is available. | How much legitimate activity would be lost, how long service may be affected, and whether the backup predates the full incident window. |
Once restored, test the affected workflow and confirm that the unwanted access remains contained before allowing the agent to operate again.
Escalate and document the incident
Contact the platform provider or your organization’s administrator or security owner when access may have been misused, when you cannot establish the full scope, or when a system needs a controlled recovery. Provide the time window, affected account and resources, agent identity if known, and preserved activity details. OpenAI’s account guidance directs users to support for suspected compromise and asks them to include details of activity they did not perform or authorize. Google advises contacting the affected resource owner and involving the security team. See OpenAI account security guidance and Google Cloud AI threat response guidance.
Quick Recap
Reduce the chance of another unwanted change
- Use a distinct identity for each agent. Grant only the roles and permissions it needs, and avoid broad production access when a narrower environment will work. Google Cloud’s guidance states: “To limit an agent’s ability to take dangerous actions, create an agent identity, and follow the principle of least privilege to grant the agent only the roles and permissions necessary to complete its tasks.” See Google Cloud MCP security documentation.
- Require deliberate approval for consequential actions. Where the agent supports approval gates, have the reviewer check the proposed action and target—not just click through. Approval reduces risk but does not guarantee correctness.
- Keep actions attributable. Maintain an audit trail that links changes to an agent instance and, where possible, to the person or task that initiated it.
- Prepare and test a recovery route. Set up backups, version history, transaction logging, or rollback infrastructure appropriate to the systems the agent can modify.
- Limit connected apps and task scope. Enable only the apps needed for the current task, avoid vague open-ended instructions, and stop the run if its behavior or the content it encounters looks suspicious. See ChatGPT agent guidance.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
