Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Short answer: install a plugin with agent-browser plugin add <ref>, inspect it with agent-browser plugin list and agent-browser plugin show <name>, then invoke it according to its declared capability. Credential providers, browser providers, launch mutators, and generic commands each have a different execution path. Treat every plugin as a separate executable, review its source and permissions, and use confirmation gates for sensitive capabilities.
The commands and precedence described here follow the live Agent-Browser configuration, commands, security, and authentication reference. Those pages can change, and they do not establish a specific Agent-Browser version.
What an Agent-Browser plugin is
A plugin extends the CLI through an external executable. It is not automatically part of Agent-Browser itself. A configured entry gives the executable a name and declares the capabilities it provides, for example:
{
"plugins": [
{
"name": "vault",
"command": "agent-browser-plugin-vault",
"capabilities": ["credential.read"]
}
]
}The package or repository is responsible for the executable. Agent-Browser supplies the protocol and capability gates, so installing a package does not by itself prove that its maintainer, dependencies, or behavior are trustworthy.
#1 Best Overall
Install a plugin from npm or GitHub
Use agent-browser plugin add with a package name, scoped package, or GitHub owner/repository reference. Plain names and @scope/name references resolve from npm; owner/repo references resolve from GitHub.
- Install an npm package and choose a local name:
agent-browser plugin add agent-browser-plugin-vault --name vault - Install a scoped npm package:
agent-browser plugin add @company/agent-browser-plugin-vault --name vault - Install from GitHub:
agent-browser plugin add org/agent-browser-plugin-cloud-browser - Use user-level scope only when appropriate:
agent-browser plugin add <ref> --globalWithout
--global, the command writes project configuration. A global entry is useful when several projects should share the same integration; a project entry keeps the dependency and behavior tied to one repository.
If the package includes a plugin.manifest, Agent-Browser can discover its name and capabilities. If it does not, supply the capability explicitly during installation:
agent-browser plugin add <ref> --capability <name>Example names in the documentation are illustrative; verify that the package exists and is maintained before installing it.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Inspect what was installed before running it
Always inspect the resulting configuration:
agent-browser plugin list
agent-browser plugin show <name>Confirm the executable command, plugin name, and capability list. Check the package’s repository, release history, dependency behavior, and requested permissions independently. A plugin that can read credentials or alter browser launch arguments deserves more scrutiny than one that performs a non-sensitive local transformation.
Where configuration lives and which value wins
Agent-Browser combines user and project settings, then applies environment variables and command-line flags. The practical precedence is:
Rank #2
- HTML CSS Design and Build Web Sites
- Comes with secure packaging
- It can be a gift option
- User configuration:
~/.agent-browser/config.json. - Project configuration:
./agent-browser.json; project values override user values and project plugin entries append after user entries. - Environment variables, which override configuration.
- CLI flags, which have the highest priority.
If the same plugin name occurs in both user and project lists, the later project entry resolves. You can replace normal config discovery with AGENT_BROWSER_PLUGINS, whose value is a JSON array of plugin entries. Keep project files reviewable so teammates can see which executable and capabilities a project adds.
Invoke the plugin through its capability
There is no universal plugin run command. Use the path that matches the capability shown by plugin show.
Free tools Windows power users keep installed
One-click scans. No signup required.
Credential providers: credential.read
A credential provider supplies login data to Agent-Browser without requiring you to put a password in a command argument:
agent-browser auth login <profile> --credential-provider <plugin> [--item <ref>]Agent-Browser says it does not save credentials returned by the provider locally. Use the vault vendor’s own login or session mechanism, or an environment-based secret outside Agent-Browser configuration; do not place vault tokens or passwords in plugin arguments.
Preserve a page with --no-navigate
When you have already opened a page and completed a stateful step—such as dismissing consent or clearing a challenge—use:
agent-browser auth login <profile> --credential-provider <plugin> --no-navigateThis option requires an active top-level HTTP(S) page. Agent-Browser compares the page and effective credential URL by scheme, host, and effective port; paths, queries, and fragments may differ. It does not promise that form submission will avoid navigation. If selectors do not match, use the login command’s per-login selector overrides.
Rank #3
Browser providers: browser.provider
A browser-provider plugin supplies a CDP WebSocket URL. Start Agent-Browser with:
agent-browser --provider <name> ...The configuration documentation discusses integrations such as AgentCore, Browser Use, Browserbase, Browserless, Kernel, and Remote Agent Browser. Their appearance in documentation is not a quality ranking or an endorsement.
Launch mutators: launch.mutate
A launch mutator appends local Chrome launch arguments, extensions, or initialization scripts. Invoke it through the normal browser launch workflow; do not try to call it as an arbitrary generic command.
Generic commands and custom capabilities
For a plugin that declares command.run or another custom capability, use:
agent-browser plugin run <name> <type> --payload '{"siteKey":"...","url":"https://example.com"}'The payload shape belongs to the plugin. The documented CAPTCHA example only demonstrates invocation syntax; it does not establish that a solver is available, suitable, or permitted for a particular site.
Secure credential and plugin workflows
Protect saved authentication profiles
The security documentation states that saved authentication profiles use AES-256-GCM. If AGENT_BROWSER_ENCRYPTION_KEY is unset, Agent-Browser generates a key on first use at ~/.agent-browser/.encryption-key. Back up that file for portability, or set the variable explicitly in your secure environment. Preserve the restrictive file permissions documented by Agent-Browser.
Rank #4
- Brand: Wiley
- Set of 2 Volumes
- A handy two-book set that uniquely combines related technologies Highly visual format and accessible language makes these books highly effective learning tools Perfect for beginning web designers and front-end developers
Add approval gates for sensitive capabilities
Use the --confirm-actions policy form to require approval before a capability runs. Examples include:
plugin:vault:credential.readplugin:cloud-browser:browser.providerplugin:stealth:launch.mutate
Choose gates based on what the plugin can access, not merely where it was downloaded from. Review changes to project configuration in code review and avoid running untrusted plugin executables in a context that contains production credentials.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →A repeatable setup checklist
- Decide the capability you actually need: credentials, hosted browser, launch customization, or a custom command.
- Evaluate the specific npm package or GitHub repository and its maintainer.
- Run
agent-browser plugin add, using--globalonly for a deliberate user-wide installation. - Open the generated project file or user config and verify the command and capabilities.
- Run
agent-browser plugin listandplugin show. - Invoke the dedicated capability path, not
plugin runby default. - For secrets, use a credential provider, encryption-key hygiene, and confirmation policies.
- Test in a non-production profile before granting access to real accounts.
Troubleshooting common failures
The plugin is not listed
Check that you installed into the configuration scope you are reading. Inspect both ~/.agent-browser/config.json and the project’s agent-browser.json, then rerun agent-browser plugin list. If AGENT_BROWSER_PLUGINS is set, it may be replacing normal discovery.
Agent-Browser cannot start the executable
Use plugin show to verify the command name and executable path. Confirm the package actually installed its binary and that the current environment can execute it. Reinstall or correct the generated entry rather than adding secrets to the command line.
The command path is rejected
Compare the declared capability with the invocation. Use auth login for credential.read, --provider for browser.provider, the regular launch flow for launch.mutate, and plugin run only for generic or custom capabilities.
--no-navigate refuses to run
Make sure the active page is top-level HTTP(S) and that scheme, host, and effective port match the credential URL. A different path or query is acceptable; a different host or port is not. If the page was not prepared first, omit --no-navigate and let the normal login flow navigate.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsLogin fields are not filled
Use the command’s per-login selector overrides and verify that the page has finished loading. A retained page does not disable navigation triggered by submitting the form.
A sensitive action runs without approval
Review the exact capability string and add the corresponding plugin:<name>:<capability> policy to --confirm-actions. The policy must match the configured plugin name and capability.
Or skip the browser setup
If your real goal is producing page screenshots rather than extending Agent-Browser, ScreenshotNeo is the first alternative to try: it removes cookie banners, newsletter popups, and chat widgets before capture; bot checks, blank pages, failed loads, timeouts, and cache hits are not billed; it offers an MCP server for AI agents; and its free plan includes 1,000 screenshots per month with no card, while paid plans start at $5 for 3,000 shots.
One GET request returns an image or PDF:
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpEquivalent clients:
import requests
r = requests.get("https://api.screenshotneo.com/v1/shot", params={"access_key": "YOUR_API_KEY", "url": "https://stripe.com"}, timeout=90)
open("shot.webp", "wb").write(r.content)const q = new URLSearchParams({ access_key: 'YOUR_API_KEY', url: 'https://stripe.com' });
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);See the complete parameter reference at ScreenshotNeo’s documentation. Sign up free for 1,000 screenshots a month with no card.
Frequently asked questions
Can one plugin provide several capabilities?
Yes, a manifest or manual entry can declare multiple capabilities. Inspect each one and apply the invocation and confirmation policy appropriate to that capability.
Should plugins be installed globally?
Only when multiple projects intentionally share the same executable and version. Project scope is easier to review, reproduce, and change independently.
Does installing a plugin store my passwords?
Credential-provider logins return credentials through the provider protocol; Agent-Browser says it does not save those returned credentials locally. Saved authentication profiles are separately encrypted as described in its security documentation.
Frequently Asked Questions
Can one plugin provide several capabilities?
Yes. A manifest or manual configuration can declare multiple capabilities; inspect and gate each capability separately.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Should plugins be installed globally?
Use global scope only for an intentionally shared integration. Project scope is generally easier to reproduce and review.
Does installing a plugin store my passwords?
Credential-provider logins return credentials through the provider protocol, and Agent-Browser says those returned credentials are not saved locally. Saved profiles are encrypted separately.
Quick Recap
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.
