Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.
Grounded in the docs - follow a source before acting on it.
Edit markdown files in Google Docs without converting them - OpenSmartRoute
Edit markdown files in Google Docs without converting them
Google Workspace now lets you open, edit, and collaborate on markdown files directly in Google Docs and Drive. You can work on .md files without changing the file type.
Key points
Open .md files natively in Google Docs without converting them.
Edit markdown files in real-time with teammates and AI agents.
View formatted previews of markdown files inside Google Drive.
Why it matters: Markdown is a common language for AI agents and developers. This change lets teams and AI tools work on the same documents without manual file conversion.
By OpenSmartRoute editorial · written through the router by llm-frontier
From Google AI developers blog - “Preview, edit, and collaborate on markdown (.md) files natively in Google Docs and Drive”
Google Workspace now lets you edit markdown files directly. You no longer need to convert them to PDF or HTML. This change removes a common step in the workflow. Developers can now work on documentation inside Google Docs. The system treats these files as native documents.
Google Docs now recognizes the .md file extension. A small badge appears in the document header. This badge tells you the file is in markdown format. You can see the raw text or the rendered preview. Clicking the preview shows how the document looks.
Markdown is a simple text format for structured content. It uses symbols like asterisks for lists and hashes for headers. LLMs read markdown easily because it is plain text. It avoids the heavy overhead of formats like Word or PDF. This makes it faster for AI tools to process information.
Google Drive has seen a big increase in markdown usage. This growth shows how important this format is for teams. People use it for notes, specs, and code documentation. The new feature lets you edit these files without leaving Drive. You can collaborate with others in real time.
The system preserves the original markdown structure. Your headers, tables, and links stay intact. Developer tools can still read the file correctly. Git repositories will see the same file content. There is no hidden conversion happening in the background.
You can now invite AI agents to edit these files. An agent can suggest changes directly in the document. Humans and agents work on the same source of truth. This reduces confusion about which version is current. Everyone sees the latest updates immediately.
Google Docs provides a familiar environment for everyone. Team members do not need special software to edit. They just use the web browser or the mobile app. The interface looks the same as editing a normal doc. This lowers the barrier for non-technical users to join.
Real-time collaboration works smoothly for multiple people. Comments appear instantly when someone types them. You can see who is editing and when they changed something. The system tracks every single revision made by anyone. This creates a clear history of all changes.
Version history is now more detailed than before. You can see edits made by AI agents specifically. You can also see edits made by human team members. The audit trail is complete and transparent. No changes are hidden from the team.
LangChain shares five core components for building internal AI agents that work inside existing tools like Slack. The system handles identity, memory, and code execution safely.
LangChain released updates to its Deep Agents harness to fix how agents search for information. New tools and patterns help agents retrieve only the exact data they need without wasting context window space.
Storage in Drive handles rich formatting well. Tables and hyperlinks display correctly in the preview. You can toggle between the view and the raw text. This flexibility helps you review formatting with a click. It also helps you check the code structure easily.
This feature bridges the gap between code and collaboration. Developers can write docs and have them reviewed instantly. Product teams can see the technical details in one place. AI tools can access the same documents for tasks. The workflow becomes much more seamless and efficient.
LLMs use markdown to preserve structured content. It keeps tables, headers, and lists readable. Without markdown, AI might lose the structure of the text. This format ensures the data remains clean for processing. It is the standard language for many AI interactions.
AI agents take on larger roles in planning and coding. They need a common language to talk to humans. Markdown serves as that shared vocabulary between people and machines. It is easy for both sides to read and understand. This makes collaboration between humans and agents much easier.
Google Docs acts as a shared bridge for the team. It brings everyone together on the same documents. Autonomous agents can work alongside human contributors. The system supports drafting, reviewing, and shipping content. It speeds up the entire technical content lifecycle.
Engineering teams benefit from this unified workspace. They can draft technical specs without switching apps. They can review them with product managers in the same thread. They can ship the final code documentation faster. The friction of manual conversions is completely gone.
You can start using this feature today. Just upload an .md file to Google Drive. Open it directly in Google Docs to begin editing. Invite your team members to collaborate in real time. Watch the version history as you work together.
This change matters for engineering and AI workflows. It saves time by removing manual conversion steps. It improves quality by keeping the file format pure. It enhances safety by allowing input guards to scan markdown. It boosts speed by letting agents edit files instantly.
OpenSmartRoute helps teams route requests to the best tools. It sends each request to the best-fit model or agent. The router scores candidates on quality, cost, speed, and safety. Teams set the weights for these scores per request.
Hard rules pin specific requests to certain models. Text with personal data stays on an on-premises model. A region or cost cap is never crossed by the router. This ensures data security and budget control.
The router learns from outcomes over time. A model that answers well gets more traffic. One that fails gets less traffic automatically. This continuous learning improves the overall system performance.
A new model is just one catalogue entry. It competes on the next request immediately. Nothing else changes in the app for the user. This makes it easy to test new models quickly.
OpenSmartRoute works with any OpenAI-compatible provider. It supports open-weight models served locally. It also works with MCP tools and A2A agents. This flexibility allows teams to choose their stack.
A savings ledger shows what each routed request cost. It compares this to what the most expensive model would have cost. Teams can see exactly how much they saved. This ledger helps track spending on AI usage.
An input guard spots prompt injection before a request leaves. It also detects personal data in the input. This protects sensitive information from leaking to public models. Security is built into the routing process.
The 'osr eval' tool measures routing accuracy on team prompts. It can fail a build if the accuracy drops too low. This ensures the router performs well before deployment.
The hosted platform keeps a models catalogue with prices. It shows public rankings built from real traffic. Teams can see which models are performing best. They can make data-driven decisions about their stack.
Start using markdown files in Google Workspace today. Upload your .md files to Drive and open them in Docs. Test the real-time collaboration with your team members. Check the version history to see how changes are tracked.
Try inviting an AI agent to edit a document. See how it suggests changes in the native markdown view. Compare the speed of editing in Docs versus converting files. Measure the time saved on manual conversions.
Review the rendered previews in Drive to check formatting. Toggle between raw text and the rich view to inspect details. Ensure your git repositories can still read the files correctly.
Check the OpenSmartRoute catalogue if you use routing. See how the router scores models on quality and cost. Configure hard rules to protect your data and budget.
Monitor the savings ledger to track your AI spending. Use the input guard to prevent data leaks in your workflow. Run 'osr eval' regularly to keep your routing accurate.
This combination of Google Docs and markdown is powerful. It brings structure and collaboration together in one place. It removes the friction that slows down modern teams. It sets a new standard for how we work with AI.
The future of documentation is native and collaborative. Teams will rely on this unified workspace more often. AI agents will become more integrated into daily workflows. The line between human and machine work will blur further.
Google Workspace is evolving to meet modern needs. It is becoming a better platform for developers and AI. The markdown feature is just the beginning of this evolution. More tools will likely follow to support this workflow.
Engineers should embrace this change to improve their productivity. Managers should encourage teams to adopt the new native features. Both groups will see benefits in speed and quality. The investment in time to learn the new workflow pays off.
The technology behind this is simple but effective. It leverages existing infrastructure to solve a common problem. It does not require new hardware or complex setups. It just works within the familiar Google ecosystem.
Teams that adopt this early will gain a competitive edge. They will be faster at drafting and reviewing technical content. They will be more efficient in their use of AI tools. The productivity gains will be noticeable in every project.
The integration of markdown into Google Docs is a significant step forward. It aligns with how modern teams actually work. It respects the needs of both humans and AI agents. It creates a smoother path for future innovations.
You can see the impact of this change in your daily work. Notice how much less time you spend on conversions. Feel the difference in collaboration speed with your team. Observe how AI agents can now contribute more effectively.
This feature is available to anyone with a Google or Workspace account. There are no extra costs to enable this capability. It is part of the standard Google Workspace offering. You do not need to request special permissions.
Just upload the file and start working. The system handles the rest seamlessly. You can focus on the content and the ideas. The technical details are managed in the background.
The story of markdown in Google Docs is one of simplicity. It solves a problem that has plagued teams for years. It brings clarity and efficiency to the workflow. It is a win for everyone involved in the process.
As you move forward, keep these points in mind. Native support means no more conversions. Real-time collaboration means faster decision making. Version history means full transparency on changes.
The benefits extend beyond just editing documents. They extend to how teams manage their knowledge base. They extend to how AI agents interact with your data. They extend to how you measure your team's productivity.
This update is a clear signal from Google. They see the value in markdown for developers. They want to make this format more accessible. They want to support the growing AI ecosystem.
Your team can now work in a more integrated way. You can build better tools and faster software. You can communicate more clearly with your stakeholders. The barrier to entry for AI adoption is lower.
The combination of markdown and Google Docs is a strong foundation. It supports the needs of modern engineering teams. It supports the needs of AI researchers and developers. It supports the needs of product managers and designers.
Everyone who uses Google Workspace can benefit from this. It is not just for developers anymore. It is for anyone who works with structured text. It is for anyone who values speed and collaboration.
The technology is ready for the next wave of AI. It is ready for the next wave of automation. It is ready for the next wave of human-AI teamwork. The stage is set for new possibilities.
You are now part of this new workflow. You are using a tool that is built for you. You are working in an environment that supports you. You are moving forward with confidence and clarity.
The future is here, and it is built on markdown. It is built on collaboration. It is built on the simple power of plain text. It is built on the trust of a unified platform.
Keep using this feature to its full potential. Keep pushing the boundaries of what is possible. Keep building better things with your team.
The road ahead is clear and open. The tools are there to help you succeed. The support is there to guide you along. The future is yours to shape.
How OpenSmartRoute helps
A team that routes requests through OpenSmartRoute can keep markdown files in their native format while still using Google Docs for collaboration. The router sends the text to the best-fit model based on the team’s catalogue weights. The team defines rules for each request, such as keeping personal data on an on-premises model or staying within a region or cost cap.
The router learns from outcomes, so a model that answers well gets more traffic and one that fails gets less. A new model is one catalogue entry and competes on the next request without changing the app. The hosted platform keeps a models catalogue with prices and public rankings built from real traffic. A savings ledger shows what each routed request cost next to what the most expensive model would have cost. An input guard spots prompt injection and personal data before a request leaves.
The team can use 'osr eval' to measure routing accuracy on their own prompts and can fail a build when it drops. OpenSmartRoute works with any OpenAI-compatible provider, with open-weight models served locally, with MCP tools and A2A agents.
Google released a new extension system for its Agents CLI tool. This allows teams to use LangChain or other frameworks while keeping the same deployment and evaluation commands.
