This wire is a sample of P0 and P1 receipts from EXVIV’s curated public set. It publishes
after an 18-hour delay. Private workspaces see the complete watchlist the same day it is reviewed.
Each row is a reviewed change with a primary source and a short before-and-after excerpt. It is proof that the machine works, not a substitute for a private watchlist.
OpenAI Codex · Codex CLI 0.161.0 makes GPT-6.1 Sol the default model in bundled and Amazon Bedrock catalogs and adds Bedrock support for multi-agent V2, Ultra reasoning, and AWS GovCloud regions. Users can sign in to MCP servers from an active terminal session using /mcp login and select microphone, speaker, and input channels for voice conversations with locally saved preferences.
BEFORE
#45544 Discourage logging full image generation results @Gan-Tu
→
AFTER
Codex CLI 0.161.0
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
Augment Code · VSCode Extension 0.904.1 introduces improved chat history saving, automatic backlog draining, and fixes for large history errors, duplicate title requests, and diagnostic failures.
BEFORE
Cosmos Week 31 Release Notes
→
AFTER
VSCode Extension 0.904.1 Release Notes
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.
Vapi · Vapi now supports a Config as Code (GitOps) capability that allows users to manage assistants, squads, tools, and tests as files in git, review changes in pull requests, and promote configurations from development to production.
BEFORE
Vapi CLI
→
AFTER
Work with Vapi from your terminal, or manage your whole configuration as code:
A documented capability change may alter product comparisons, roadmap priorities, or customer objections.
Deepgram · Improved Nova-3 Medical English models for batch and streaming have been released with fewer word errors and medical entity errors. The model improvements are live for all users without requiring changes to API requests.
BEFORE
August 26, 2026
→
AFTER
October 7, 2026
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
Ada · Amazon Connect handoff now supports IAM role access secured with an External ID, which is the recommended option for new setups. Users can forward Amazon Connect queue events to the integration to record the queue and entry time as conversation meta variables.
BEFORE
Suppress the CSAT survey with rules
→
AFTER
Amazon Connect handoff supports IAM role access
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
Zed · The release adds the ability to preview JSONL and NDJSON files as tables, accessible via a keyboard shortcut. A new setting, agent.max_idle_retained_threads, was added to control how many idle agent threads remain loaded.
BEFORE
This week's release includes tabular data previews for CSV, TSV, PSV, and SSV files with sortable and resizable columns, value-based row filtering, and right-click copying; new Git blame and stashing actions; and lower memory use when opening large files.
→
AFTER
This week's release includes the ability to preview JSONL and NDJSON files as tables.
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.
OpenAI Codex · ChatGPT for iOS 1.2026.272 adds inline page previews, direct Codex task link support, a redesigned approval picker, a unified Plugins menu, faster streaming, persistent computer selection during reconnection, and improved worktree support.
BEFORE
#46310 Defer environment selection changes until the next turn @sayan-oai
→
AFTER
ChatGPT for iOS 1.2026.272
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.
Cursor · Cursor iOS app now allows users to see and reply to local agents running on their computer. Remote Control is enabled by default for all users except those in Enterprise organizations.
BEFORE
Aug 19, 2026 · Changelog
→
AFTER
Oct 6, 2026 · Changelog
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.
Deepgram · Nova-3 Medical now supports multilingual transcription in Dutch, English, French, German, Hindi, Italian, Japanese, Portuguese, Russian, and Spanish via the language=multi parameter. Nova-3 Medical Multilingual is available through the API for hosted and self-hosted customers using the model=nova-3-medical and language=multi parameters.
BEFORE
_Showing the 20 most recent of 228 entries. Append `/llms.txt` to the changelog URL for the complete index._
→
AFTER
October 6, 2026
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
Ada · Access to Voice Proactive Outreach requires contacting the Ada team. The Conversations API places calls over the customer's own outbound SIP trunk and detects answering machines.
BEFORE
Dashboard tables remember your sort
→
AFTER
Initiate phone calls with your AI Agent using Voice Proactive Outreach
A new security, governance, or deployment capability can open larger accounts and reduce procurement friction.
Mistral AI · Mistral enables customers to customize frontier open-weight models with Forge and deploy them with Studio on-premises, in their VPC, or in Mistral's cloud.
BEFORE
Mistral AI | #1 in Sovereign AI
→
AFTER
Mistral | #1 in Sovereign AI: Frontier Open Models You Own
A new security, governance, or deployment capability can open larger accounts and reduce procurement friction.
Bubble · Bubble reduced the annual billing prices for the Starter, Growth, and Team plans to $29, $119, and $349 per month, respectively. The Enterprise plan call-to-action changed from 'Talk to sales' to 'Get pricing' while retaining the same contact-sales destination.
BEFORE
Talk to sales
→
AFTER
Get pricing
A verified pricing or limit change may affect competitive positioning and customer comparisons.
Langfuse · Langfuse introduced a new chart that displays organizational units over time grouped by project or type, with CSV export available on the Billing page for Cloud and a new Usage page for self-hosted deployments.
BEFORE
Langfuse MCP: now with Observations, Metrics, Scores, Comments and more
→
AFTER
See which projects drive your usage
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.
OpenAI Codex · Auto-review safety checks are now free and do not count toward usage limits for signed-in ChatGPT accounts, whereas they previously used GPT-5.6 Luna. Auto review is no longer explicitly listed as a supported activity type in the general rate applicability statement.
BEFORE
These rates apply to supported ChatGPT Work and Codex activity, including local tasks, cloud tasks, automations, code review, auto review, and delegated workers. Charges are based on the model used and the actual input, cached input, and output tokens consumed.
→
AFTER
These rates apply to supported ChatGPT Work and Codex activity, including local tasks, cloud tasks, automations, code review and delegated workers. Charges are based on the model used and the actual input, cached input, and output tokens consumed.
A packaging or usage-policy change can alter effective price, product eligibility, and customer comparisons.
Trigger.dev · Trigger.dev v4.7.3 requires explicit trigger permissions for session creation, adds chat session tags, and includes six server improvements for faster trace loading and Logs page performance. The Concurrency page now charts scheduling delay, backlog, and throttling for every queue and concurrency key, with live overrides and archiving for unused queues.
BEFORE
Trigger.dev v4.5.13
→
AFTER
Trigger.dev v4.7.3
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
OpenAI Codex · ChatGPT for iOS 1.2026.267 introduces Home Screen and Lock Screen widgets that display recent tasks and track remaining usage and reset times. The update adds a setting to keep the keyboard open after sending a message and ensures canceling a queued prompt returns recoverable text and attachments to the composer.
BEFORE
#46487 Keep TUI exploration grouped across reasoning and nonzero exits @etraut-openai
→
AFTER
ChatGPT for iOS 1.2026.267
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.
Ada · Users can now choose which types of sensitive images their AI Agent withholds, with toggles for government ID and credential documents enabled by default. Payment cards are always withheld, and card numbers read from any image are always redacted.
BEFORE
Manage change sets from the dashboard
→
AFTER
Choose which sensitive image types the AI Agent reads
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
Gorgias · The unresolved incident of elevated latency in Helpdesk and Ticket Views has been resolved, and all systems are now operational with no incidents reported.
BEFORE
Subscribe to updates for Elevated latency in Helpdesk and Ticket Views via. You'll receive.
→
AFTER
All Systems Operational
A material reliability change can affect buyer confidence, renewal risk, and competitive displacement opportunities.
OpenAI Codex · Codex CLI 0.160.1 was released on 2026-10-05 with a bug fix that preserves SYSTEMROOT, TEMP, and TMP environment variables when launching remote stdio MCP servers.
BEFORE
#46505 Support catalog parameter schemas for Multi-Agent V2 tools @rhan-oai
→
AFTER
2026-10-05 Codex CLI 0.160.1 $ npm install -g @openai/codex@0.160.1 View details Bug Fixes Preserve SYSTEMROOT, TEMP, and TMP when launching remote stdio MCP servers with explicitly configured remote environment variables, allowing Unix hosts to retain the Windows executor's startup environment. Changelog #51121: Backport Windows remote MCP environment preservation to 0.160. Full release on Github
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
PolyAI · The service is aligned with National Cyber Security Centre Guidelines and conducts regular audits and continuous third-party penetration testing to ensure security and compliance.
BEFORE
Trusted by
→
AFTER
Amazon Web Services
A new security, governance, or deployment capability can open larger accounts and reduce procurement friction.
Lovable · The company is hiring an AI Ops Engineer focused on the Marketing function. The company is hiring a Staff Software Engineer for the Enterprise Platform.
BEFORE
IT Specialist
→
AFTER
AI Ops Engineer (Marketing)
A sustained hiring change can indicate where the competitor is investing before the strategy is formally announced.
n8n · The Anthropic node's Message operation now supports a Prompt Caching option with 5-minute or 1-hour durations to reduce costs for repeated tool-call loops and large system prompts. Agents can now pay for web search using Gateway credits when Brave Search is served through the AI Gateway, eliminating the need for a personal Brave API key.
BEFORE
Add option to select eval suite on manual workflow dispatch: The internal AI evals GitHub Actions workflow now lets maintainers choose which evaluation suite to run when manually triggering it: both suites (default), spec (pairwise) only, or matrix (LLM judge) only. Automatic runs triggered by pushes to master continue running both suites as before. This is an internal CI and testing tool change and does not affect…
→
AFTER
n8n 2.42 Anthropic node gains prompt caching support, plus 13 other features
A material reliability change can affect buyer confidence, renewal risk, and competitive displacement opportunities.
Trigger.dev · Users can now debug, investigate, and generate reports on their runs by chatting with Ask Trigger from their dashboard. Preview branches can now be configured to automatically archive themselves after a specified period of inactivity.
BEFORE
Bulk actions from the SDK
→
AFTER
Introducing Ask Trigger to your dashboard
A new release or deprecation can change feature parity, migration risk, and the next competitive response.
Ada · Users can now set availability rules on knowledge sources for the Zendesk App, Salesforce Knowledge App, Contentful, Dixa, Gladly, ServiceNow, Helpjuice, and Microsoft Dynamics. Availability rules only to articles added after they are saved, while existing articles retain their own rules.
BEFORE
Platform API: daily rate limits removed on several endpoints, new rate limit headers
→
AFTER
Availability rules for integration knowledge sources
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.
n8n · n8n 2.41 adds @mention support to Microsoft Teams messages, allowing users to resolve and send real notification chips via a new user picker. The release makes the Databricks Chat Model node visible in the node picker and enables MiniMax nodes to fetch available models dynamically from the API.
BEFORE
n8n 2.40 Add Microsoft Dataverse core node, plus 31 other features
→
AFTER
n8n 2.41 Add mentions to Microsoft Teams messages, plus 7 other features
An ecosystem expansion can add distribution, increase switching costs, or close an adoption gap.