AI.info
AI Tools Updates — Page 27
What changed in the AI tools this week — dated, from the makers’ own changelogs.
- Ask Sage — Ask Sage v4.0.1
Reduces timeouts on long responses and gives Anthropic's newest government models more room for longer answers.
- Devin — November 2, 2026
Read-only access is clearer, Fix with Devin links to the relevant session or rows, and code scans support up to 1,000 repositories.
- Devin — October 30, 2026
Markdown and patch files open as rendered previews; Files tabs, multiline secret editing, and GitHub file links are improved.
- incident.io On-call — Summary and update guidance
Choose how summaries and updates are written across your organization.
- incident.io Incident Response — Summary and update guidance
Choose how summaries and updates are written across your organization.
- Lokalise — Save Machine Translation into Translation Memory in Tasks and the API
You can now save Machine Translation straight into your Translation Memory when you work with Tasks and through the public API.
- Lokalise — Safely remove custom filters, even when Workflows use them
Deleting a custom filter now works in the editor, in task scope selection and in the workflow trigger.
- Lokalise — Analytics has new Workflows and Automations dashboards
Two new dashboards on the Usage tab of Analytics give you a full view of how your Workflows and Automations run and what they deliver.
- Claude Code — 2.1.317
- MindBridge AI Auditor — MindBridge brings financial risk intelligence into everyday AI assistants with new MCP launch
New MCP capabilities let finance and audit teams explore MindBridge analyses, investigate risk, and access explainable insights in AI workflows.
- Hermes Agent — Hermes Agent v0.22.7
Added opt-in lead routing and cron posts for Desktop Group Chats, plus room purpose and shared notes.
- Devin Desktop — November 2, 2026
Code scans support up to 1,000 repositories; read-only access is clearer, Fix with Devin links to sessions, and auto-scan controls are added.
- Devin Desktop — October 30, 2026
Markdown and patch files open as previews; folder navigation, multi-line secrets, and GitHub file-link navigation are improved.
- Intercom Fin — Clear reasons and filters for emails in the Spam inbox
The Spam inbox now shows reasons for emails and offers filters to find them.
- Intercom Fin — Let Fin Handle Order and Product Questions in Shopify, WooCommerce and BigCommerce
Fin can handle order and product questions in Shopify, WooCommerce, and BigCommerce.
- Intercom Fin — Your News site now has an RSS feed
News sites that allow search engine indexing now publish an RSS feed of their 50 most recent updates.
- Intercom Fin — Let Fin Manage Subscriptions in Stripe, Chargebee, Paddle and Recurly
Fin can manage subscriptions in Stripe, Chargebee, Paddle, and Recurly.
- Fireworks AI — New Serverless mode: Ultra-fast (Preview)
Ultra-fast is a new high-speed Serverless mode, now in Preview, for highly latency-sensitive applications.
- GLM Coding Plan — 3.17.2 Released Nov 3, 2026
Added the official ZCode Sites plugin with one-click website deployment; Added unread task filtering; Added an MCP configuration interface (draft)
- Composio — py@0.31.0
Released Python SDK 0.31.0.
- Composio — @composio/core@0.28.0: Release: update version (#5070)
POST and PATCH requests are sent once instead of retried; other requests keep their retries.
- Composio — @composio/slim@0.28.0: Release: update version (#5070)
POST and PATCH requests are sent once instead of retried; other requests keep their retries.
- Palantir Foundry — DeepSeek V4.1 Flash (1017) support in AIP through Fireworks
AIP adds support for DeepSeek V4.1 Flash (1017) through Fireworks, including text and image inputs and tool calling.
- Palantir Foundry — Bring public web information into your AIP workflows
Web search lets OpenAI models in AIP find information from public web sources and return answers with citations.
- Palantir Foundry — Pipeline builder defaults to faster pipelines, now with LLM support
Faster pipelines are now the default in Pipeline Builder, with support for Use LLM nodes and Text to Embeddings nodes.
- Voiceflow — Messages sent mid-turn interrupt your agent when possible
A message that arrives while your agent is busy is held on v4 socket and voice conversations, then interrupts the turn or runs when it ends.
- Voiceflow — Close an evaluation's log by clicking its row again
Clicking an open row again in the Results table closes an Evaluation log.
- Voiceflow — Read evaluation release cards in full over the criteria table
A release card opens in full over the criteria table when you hover a release marker.
- Retool Agents — Retool on Google Cloud Marketplace
Retool is available on Google Cloud Marketplace, where customers can accept a private offer and bill through Google Cloud.
- Retool Agents — App testing
App testing makes the agent write and run tests when it adds or changes app behavior.
- OpenHands Cloud — openhands: v0.96.0
Added admin console toggles for Agent Profiles and Model Routing, on by default.
- SambaNova DataScale — Supermicro Will Build and Sell SambaNova-Powered Inference Systems
Supermicro will deliver and support inference systems pairing NVIDIA GPUs with SambaNova SN50 RDUs, rolling out in phases.
- Bulbul — Introducing Sarvam Prism
Introduced Sarvam Prism.
- Kaiber (Superstudio) — Music Video Montage 2.0
Music Video Montage 2.0 adds stronger character, scene, and style consistency, plus Max, Turbo, and Lite model options.
- Snowflake Data Cloud — Snowflake CoCo for Excel and Snowflake CoCo for Google Sheets (Preview)
Snowflake CoCo for Excel and Snowflake CoCo for Google Sheets
- Snowflake Data Cloud — Cortex AI Gateway inference, observability, and cost management (General availability)
Cortex AI Gateway inference, observability, and cost management
- Snowflake Data Cloud — CoCo Desktop v1.24.0
- Snowflake Data Cloud — Delta Direct: Structured type evolution (General availability)
Delta Direct: Structured type evolution
- Aftershoot Editing — Your Aftershoot Founder Membership: How It Compares This Black Friday
Founder Membership costs $499 for the first year at Black Friday pricing, then stays $499/year while the subscription is active.
- Ursa — 2026.11.03 Realtime SaaS
Realtime sessions are now less likely to become unresponsive when transcription falls behind the audio being sent.