Skip to content

Remote agents

Remote agents are curated, cloud-hosted agent definitions maintained by the TeXRA team. They provide specialized research capabilities and domain expertise, receiving continuous prompt and configuration updates without requiring a local extension update.

What are remote agents?

Remote agents are pre-configured agents hosted in the cloud that you can run directly from TeXRA. Unlike built-in agents, they:

  • Receive updates automatically, without an extension update
  • Provide specialized capabilities for specific research domains
  • Load on demand, so the extension stays small

Getting started

1. Sign in to TeXRA

Remote agents require a TeXRA account:

  1. Open the Command Palette (Ctrl+Shift+P / Cmd+Shift+P)
  2. Run TeXRA: Sign In
  3. Choose a sign-in method (GitHub or Google)
  4. Complete the sign-in flow in your browser
  5. Return to VS Code once signed in

TIP

To check your sign-in status at any time, run TeXRA: View Profile from the Command Palette.

2. Browse available agents

Once signed in, browse remote agents from the Agents tab:

  1. Open the Command Palette
  2. Run TeXRA: Show Agents to open the Agents tab in Settings
  3. Select a remote agent in the list to see its details
  4. Select Show in agent selector to add it to your agent selector

Your email and access level sit on the Account & Usage tab (TeXRA: View Profile).

The selected agent appears in the main TeXRA view alongside your built-in agents.

3. Use remote agents

Remote agents work like built-in agents:

  1. Open the TeXRA view (select the TeXRA icon in the Secondary Side Bar)
  2. Select your remote agent from the agent dropdown
  3. Choose your input file and model
  4. Select Execute to start the agent

INFO

Remote agents carry a cloud icon (☁️) in the agent selector to set them apart from local agents.

The agent dropdown is a single flat list with no Local/Remote headers; remote agents are told apart only by the trailing cloud icon (tooltip: Remote agent, prompts loaded from cloud), and a Browse all agents… entry at the end opens Settings → Agents:

Agent
search
polish
correct
searchin use
simplifier
orchestrator

The agent dropdown: one flat list where remote agents (search, simplifier, orchestrator) carry a cloud marker that sets them apart from local agents like polish and correct.

None of this is VS Code-only. The whole loop works from a terminal too:

texra login
$texra login
Pick a provider: GitHub / Google → browser opens
Signed in as ada@lab.edu.
$texra agents show search
name: search
category: toolUse
source: remote
$texra chat --agent search
runs like any local agent — same chat, same run history

Sign in once and remote agents resolve by name everywhere: agents show reports source: remote, and texra chat --agent search runs it like any local agent.

Remote agent access

Signed-in researchers, academics, and technical writers can use specialized remote agents for academic research and professional writing. Signing in only unlocks the hosted agent catalog. Remote agents still run on your own credential, the same provider API key or subscription as your built-in agents. Access includes:

  • Specialized domain agents: Agents tailored for specific research fields (mathematics, computer science, physics, and others)
  • Advanced reasoning capabilities: Multi-step analysis and complex document processing
  • Beta features: Early access to experimental agents and capabilities
  • Priority improvements: Your feedback directly influences agent development
Specialized domain agents

Agents tailored for specific research fields.

mathematicsCSphysics
Advanced reasoning

Multi-step analysis and complex document processing.

multi-step
Beta featuresEarly access

Early access to experimental agents and capabilities.

Priority improvements

Your feedback directly influences agent development.

your feedback

What remote agent access gets you: domain-specific agents, advanced multi-step reasoning, early access to beta agents, and a direct line into how agents evolve.

Current remote agents include search (literature discovery), simplifier (code and writing simplification), and orchestrator (multi-agent coordination). Read the remote agents section of the built-in agents guide for details on each.

Different agents may be available to different research groups, depending on their domain and needs.

Requesting access

Remote agent access is available to active researchers, academics, and technical writers. To request it:

  1. Sign in to TeXRA using your institutional or professional email
  2. Contact us at contact@texra.ai with:
    • Your name and affiliation
    • A brief description of your research area
    • How you plan to use remote agents

We review each request and grant access within 1-2 business days.

Free for researchers

TeXRA supports academic research. Remote agent access is free for qualifying researchers and students: there is no charge for the hosted agent catalog. Model calls made by those agents are still billed to your own provider account or subscription.

Managing your account

View your profile

To check your account status:

  1. Open the Command Palette
  2. Run TeXRA: View Profile
  3. View your email and access level on the Account & Usage tab

Run TeXRA: Show Agents to browse the remote agents available to you.

Sign out

To sign out of your TeXRA account:

  1. Open the Command Palette
  2. Run TeXRA: Sign Out
  3. Confirm sign out

Your local agents continue to work after signing out. Only remote agents become unavailable until you sign in again.

Frequently asked questions

Do I need to sign in to use TeXRA?

No. TeXRA works fully offline with built-in agents. Sign in only if you want remote agents from the cloud.

Are remote agents slower than built-in agents?

Remote agents perform similarly to built-in agents. The agent configuration is loaded from the cloud, but the model runs with your configured API keys, as with built-in agents.

Can I use remote agents offline?

No. Remote agents require an internet connection to load their configurations. For offline work, use built-in agents instead.

What happens to my documents?

Your documents are never uploaded to TeXRA servers. Remote agents only load their configuration (prompts and settings) from the cloud. All document processing happens locally using your own API keys.

Can I create my own remote agents?

Currently, remote agents are maintained by the TeXRA team. If you have ideas for new agents or want to share your custom agents with the community, contact us at contact@texra.ai.

How do I disable remote agents?

Remote agents are available only while you are signed in. To stop using them, sign out; your built-in agents keep working. (In the CLI, run texra auth logout; in the VS Code extension, use the sign-out action in the account menu.)

Need help?

If you run into issues with remote agents or have questions about access:

We typically respond within 24-48 hours.