docs: add preview release notes by pierrevalade · Pull Request #391 · anotherai-dev/anotherai

pierrevalade · 2025-10-02T14:58:25Z

Summary

Added preview release notes documentation

🤖 Generated with Claude Code

🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

vercel · 2025-10-02T14:58:33Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
anotherai	Ready	Preview	Comment	Oct 6, 2025 8:19pm
anotherai-docs	Error			Oct 6, 2025 8:19pm

pierrevalade · 2025-10-02T15:01:07Z

docs/content/docs/releases/preview.mdx

+
+We've also published a note about how we have secured the `query_completions` tool [here](/security#sql-query-tool-security) from malicious use. We welcome more feedback on our approach via our [Slack](https://join.slack.com/t/anotherai-dev/shared_invite/zt-3av2prezr-Lz10~8o~rSRQE72m_PyIJA)
+
+## Some (current) limitations.


@guillaq what do you think? should we add more limitations?

No I can't think of any. Maybe the fact that we have focused our support of OpenAI SDK for now ?

- Changed bullet points to en-dashes in preview.mdx - Updated index.mdx with preview announcement callout 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

anyacherniss · 2025-10-02T17:49:50Z

docs/content/docs/releases/preview.mdx

+description: Public preview
+---
+
+Today we're introducing a public preview of **AnotherAI**, a MCP server designed for AI engineering tasks that includes a set of tools that enables your AI assistant (such as Claude Code, Cursor, etc.) to:


Suggested change

Today we're introducing a public preview of **AnotherAI**, a MCP server designed for AI engineering tasks that includes a set of tools that enables your AI assistant (such as Claude Code, Cursor, etc.) to:

Today we're introducing a public preview of **AnotherAI**, a MCP server designed for AI engineering tasks that includes a set of tools that enables your AI assistant (such as Claude Code, Cursor, ChatGPT, etc.) to:

Include ChatGPT? Or not something we want to highlight anymore

I'm not yet 💯 convinced that ChatGPT works well with MCP servers as of now, but I need to test more.

docs/content/docs/releases/preview.mdx

anyacherniss · 2025-10-02T17:56:54Z

docs/content/docs/releases/preview.mdx

+
+## Compare models' performance, price, and latency.
+
+AnotherAI's MCP server exposes tools that let your AI assistant access over 100 models, and compare their performance, price and latency. In our own tests, we've found that models like Opus 4 are very good at reviewing work from other models, and the latest development of longer context windows (Sonnet 4.5 supports up to 1M tokens, Gemini has models with 2M context window) makes it possible to compare more parameters (models and prompts) and agents with longer inputs.


This paragraph is clear and informative, but it would be ever stronger if it clearly framed why the user should care about this, ie. "What problem is model comparison - specifically AI-drive model comparison - solving for users?", "How will this make your [the reader's] product better/life easier?"

do you think that people using LLMs are not aware yet of the benefits of comparing different models/prompts? that does not seems to be my experience with customers.

To answer your specific question: all users I've spoken to are aware of the benefits, but that's because they're already WorkflowAI users (who have seen the benefit)

In most (all?) cases, I think being explicit about benefits and why readers should care is a relevant part of marketing a product, hence the feedback I left. But if you feel that all potential users will already be experienced enough with building LLM agents to understand the benefits of comparing models (and experienced LLM/agent users are the only user base we're interested in) then this can be disregarded.

docs/content/docs/releases/preview.mdx

Co-authored-by: Anya <75702826+anyacherniss@users.noreply.github.com>

Update language to be more concise and accurate regarding context windows. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

guillaq

Nice !

The only critique i would have is that just reading this I don't understand how AnotherAI can have access to my production data and so it might seem too good to be true?
Maybe we should push on implementing the import completion endpoint ?

docs/content/docs/releases/preview.mdx

🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

pierrevalade · 2025-10-06T12:44:32Z

The only critique i would have is that just reading this I don't understand how AnotherAI can have access to my production data and so it might seem too good to be true?

I've added a section FAQ to address this question at the end.

- Fix grammar and consistency issues - Add FAQ section about MCP access to completions - Add note about alternative observability approach - Expand prompt examples and limitations section 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

pierrevalade · 2025-10-06T12:45:12Z

Maybe we should push on implementing the import completion endpoint ?

I really like the import completion endpoint but I think we should ship ASAP and the endpoint isn't ready/tested.

guillaq

Sounds good

guillaq · 2025-10-06T12:47:54Z

docs/content/docs/releases/preview.mdx

+
+We've also published a note about how we have secured the `query_completions` tool [here](/security#sql-query-tool-security) from malicious use. We welcome more feedback on our approach via our [Slack](https://join.slack.com/t/anotherai-dev/shared_invite/zt-3av2prezr-Lz10~8o~rSRQE72m_PyIJA)
+
+## Some (current) limitations.


No I can't think of any. Maybe the fact that we have focused our support of OpenAI SDK for now ?

anyacherniss

🚀

jacekzimonski · 2025-10-07T14:12:43Z

@anyacherniss @pierrevalade
Newest version of the preview.mdx was added to the web from this repo in PR:
#444

And is used as the Home Page content.

I think we can close this PR

docs: add preview release notes

b8dc1ca

🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

pierrevalade self-assigned this Oct 2, 2025

vercel bot deployed to Preview – anotherai October 2, 2025 15:00 View deployment

pierrevalade commented Oct 2, 2025

View reviewed changes

vercel bot deployed to Preview – anotherai-docs October 2, 2025 15:01 View deployment

docs: update preview announcement and index page

94d5dca

- Changed bullet points to en-dashes in preview.mdx - Updated index.mdx with preview announcement callout 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

vercel bot deployed to Preview – anotherai October 2, 2025 15:11 View deployment

vercel bot had a problem deploying to Preview – anotherai-docs October 2, 2025 15:12 Failure

anyacherniss reviewed Oct 2, 2025

View reviewed changes

Apply suggestions from code review

d51c979

Co-authored-by: Anya <75702826+anyacherniss@users.noreply.github.com>

vercel bot deployed to Preview – anotherai October 2, 2025 22:52 View deployment

vercel bot had a problem deploying to Preview – anotherai-docs October 2, 2025 22:53 Failure

docs: refine preview release notes wording

6e15d68

Update language to be more concise and accurate regarding context windows. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

guillaq reviewed Oct 3, 2025

View reviewed changes

docs/content/docs/releases/preview.mdx Show resolved Hide resolved

docs/content/docs/releases/preview.mdx Show resolved Hide resolved

docs/content/docs/releases/preview.mdx Show resolved Hide resolved

docs/content/docs/releases/preview.mdx Outdated Show resolved Hide resolved

docs: add free credits mention to preview release notes

6a1ba0a

🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>

pierrevalade marked this pull request as ready for review October 6, 2025 12:44

vercel bot deployed to Preview – anotherai October 6, 2025 12:46 View deployment

vercel bot had a problem deploying to Preview – anotherai-docs October 6, 2025 12:46 Failure

guillaq approved these changes Oct 6, 2025

View reviewed changes

anyacherniss approved these changes Oct 6, 2025

View reviewed changes

Merge branch 'main' into pierre/add-preview-release-notes

6ac7be4

vercel bot deployed to Preview – anotherai October 6, 2025 20:19 View deployment

vercel bot had a problem deploying to Preview – anotherai-docs October 6, 2025 20:19 Failure


		We've also published a note about how we have secured the `query_completions` tool [here](/security#sql-query-tool-security) from malicious use. We welcome more feedback on our approach via our [Slack](https://join.slack.com/t/anotherai-dev/shared_invite/zt-3av2prezr-Lz10~8o~rSRQE72m_PyIJA)

		## Some (current) limitations.

	Today we're introducing a public preview of AnotherAI, a MCP server designed for AI engineering tasks that includes a set of tools that enables your AI assistant (such as Claude Code, Cursor, etc.) to:
	Today we're introducing a public preview of AnotherAI, a MCP server designed for AI engineering tasks that includes a set of tools that enables your AI assistant (such as Claude Code, Cursor, ChatGPT, etc.) to:


		## Compare models' performance, price, and latency.

		AnotherAI's MCP server exposes tools that let your AI assistant access over 100 models, and compare their performance, price and latency. In our own tests, we've found that models like Opus 4 are very good at reviewing work from other models, and the latest development of longer context windows (Sonnet 4.5 supports up to 1M tokens, Gemini has models with 2M context window) makes it possible to compare more parameters (models and prompts) and agents with longer inputs.

Conversation

pierrevalade commented Oct 2, 2025

Summary

Uh oh!

vercel bot commented Oct 2, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

guillaq left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

pierrevalade commented Oct 6, 2025

Uh oh!

pierrevalade commented Oct 6, 2025

Uh oh!

guillaq left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

anyacherniss left a comment

Choose a reason for hiding this comment

Uh oh!

jacekzimonski commented Oct 7, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

vercel bot commented Oct 2, 2025 •

edited

Loading