Releasesby GlyphOctober 2026
See how AI answers describe your brand for one prompt GET /projects/{projectId}/visibility/prompts now returns sentimentScore in each prompt's metrics : how that prompt's answers describe your brand, from 0 to 100 (positive counts 100, neutral 50, negative 0).
Also shipped Oct 19, 2026Searchable in October 2026 →
Training Upcoming Serverless Training deprecation: DeepSeek V4 Flash (0731) On November 2, 2026 , DeepSeek V4...
Training Upcoming Serverless Training deprecation: DeepSeek V4 Flash (0731) On November 2, 2026 , DeepSeek V4 Flash (0731) will be removed from the shared Serverless Training pool. This change applies only to Serverless Training. If you train DeepSeek V4 Flash (0731) on Serverless Training, migrate to DeepSeek V4.1 Flash 1017 before November 2, 2026.
Cloud Workload Security is now available in Beta
Cloud Workload Security (CWS) is now available in Beta. Apply policy controls to cloud API requests before they reach your cloud provider. Define policies with request filters and Rego checks. For example, block requests that would make an R2 bucket public. Use workloads and policy sets to target your cloud environments.
Copilot model policy API for enterprise, orgs, and teams
You can now use the REST API to retrieve and update Copilot model policies for enterprises, enterprise teams, enterprise apps, and organizations. What’s new GET and PATCH endpoints are now generally available for model policies at four levels: Enterprise: /enterprises/{enterprise}/copilot/model-policies Enterprise teams...
GitHub Copilot models deprecated on October 19, 2026
As of today, October 19, 2026, we have deprecated the following models across most GitHub Copilot experiences (including Copilot Chat, inline edits, ask and agent modes, and code completions). These deprecations were announced on September 18, 2026 .
Postgres wait events, now in Query Insights
Query Insights for ClickHouse Managed Postgres now includes wait event breakdowns, blocked-session charts, and lock-wait charts. These metrics are also available in Prometheus.
Send OpenTelemetry traces to Agent Observability, no SDK required
You can now send OpenTelemetry traces straight to Agent Observability, without the SDK. If your agents already emit OpenTelemetry generative AI spans, from your framework, an instrumentation library, or spans you write yourself, keep your existing route through Grafana Alloy or an OpenTelemetry Collector and opt in your AI spans with a single attribute.
More from SearchableEvery Searchable release →
October 2026
Ask Searchable from Slack and Microsoft Teams Searchable's MCP server now accepts sign-ins from Slackbot and Microsoft Copilot Studio, and lets Microsoft 365 Copilot load its interactive cards. The new Slack and Teams guide covers Claude Tag, Slackbot, Copilot Studio and Microsoft 365 Copilot.
October 2026
Create API keys that can make changes Creating an API key or an MCP access token now offers Allow write access . It's off by default, so new keys stay read-only. Switched on, the key carries write and every capability, so it can use the endpoints and MCP tools that change a project: prompts, topics, competitors, the brand profile, opportunities, projects...
October 2026
MCP tool calls are now rate limited MCP tool calls are limited to 600 per minute per API key or OAuth connection. Over the limit, a tool returns rate_limited with retryable: true and retryAfterSeconds ; wait that long, then retry.
October 2026
Filter competitor data to Direct or SERP competitors Competitor rows now say whether each competitor is direct (one you track on purpose) or serp (one found in AI answers), the same split as the app's Competitors list.
October 2026
Track a state when you edit prompts in bulk POST /projects/{projectId}/prompts/sync and the update_prompts MCP tool now read a location's state . A state with no city is tracked as that state. An unrecognised state, or an ambiguous US city, is left out with a note. A state the sync could not check leaves that prompt's locations as they are, with a note.
October 2026
A city the sync could not check is no longer tracked country-wide When POST /projects/{projectId}/prompts/sync or the update_prompts MCP tool can't check a new city, because the location lookup failed or didn't answer, it now leaves that prompt's locations as they are, with a note that says to try again, as it already did for a state.
October 2026
Mention counts that follow your window and filters GET /projects/{projectId}/competitors now returns mentionsInWindow beside mentions on every competitor row and on the brand block.
September 2026
Read whole AI answers, not the first 2,000 characters GET /projects/{projectId}/prompts/{promptId}/responses and the get_prompt_answers MCP tool used to cut every answer's text at 2,000 characters, which clipped about half of all ChatGPT answers and nearly every Copilot answer. text now carries the full answer.