August 2026 Summaries

groundcover’s August 2026 dashboard update introduces agent-assisted dashboard creation from uploaded sketches, allowing users to turn whiteboard drawings or mockups into working dashboards without needing prior gcQL or widget-library knowledge. New visualization capabilities include thresholds, tree maps, conditional formatting, expanded Y-axis settings, multi-query formulas, and time-offset comparisons, while operational improvements add search and tagging, version history and restoration, stale-edit protection, cross-source variables, widget-specific sharing links, PDF and PNG exports, and synchronized chart crosshairs. The release also adds an APM dashboard pack and expands gcQL as a common query language for logs, metrics, traces, events, entities, and monitors, with cross-table queries, monitor code mode, multi-source joins, grok filtering, and stronger trace search. groundcover emphasizes that agent-generated queries remain visible, editable, and reusable, positioning dashboards as collaborative tools for both automated agents and human operators.

groundcover reports that its partner ecosystem has gained momentum less than a year after launch, with partner-sourced pipeline growing eightfold to exceed 10% of total pipeline and referrals from AWS and Google Cloud contributing more than 5%. The program targets customer concerns around telemetry costs, infrastructure complexity, AI workload observability, operational risk, and hybrid or air-gapped deployments, positioning groundcover’s rapid deployment and low management overhead as opportunities for partners to provide services and modernization support. Its network now includes resellers and service partners across the UK, Netherlands, Southern Europe, India, and Latin America, ranging from boutique consultancies to larger firms. Following its Series C funding, groundcover plans to expand partner investments through self-paced technical and sales certifications in Q3 2026, a Market Development Fund program in early 2027, and deeper co-selling, co-marketing, and co-building efforts with AWS and Google Cloud across multiple regions.

Groundcover has introduced first-class OAuth connectors for Slack, Linear, Notion, Cursor Cloud Agents, and Pylon, enabling its observability agent to combine telemetry with organizational knowledge and act within the tools teams already use. Administrators can enable connectors organization-wide and configure per-tool permissions, including approval gates for actions such as creating or modifying Linear tickets. In Slack, users can ask the agent to investigate alerts, retrieve logs, traces, RUM sessions, and dashboards, share findings in threads, and create Linear issues containing root-cause analysis and suggested code fixes. The post illustrates this workflow with a checkout outage caused by an active outage-simulation flag in a payments service, which the agent traced through telemetry and documented in a ticket. Groundcover also supports reusable incident-investigation skills that automate monitor review, Slack acknowledgments, parallel analysis of telemetry, GitHub code searches, ticket creation, and follow-up messages while offering fallbacks when integrations are unavailable. The company frames these integrations as a way to create a feedback loop in which human discussions, dashboards, tickets, and documentation continuously improve the agent’s future investigations, while complementing rather than replacing coding agents.

Groundcover’s July 2026 updates focus on making its observability agent more context-aware and accessible across existing workflows through OAuth-based connectors for tools such as Slack, Linear, Notion, Pylon, and Cursor Cloud Agents, with configurable permissions, approval gates, RBAC, budgets, reusable skills, and infrastructure-as-code support. The company also rebuilt monitor issue management with incident timelines, silence and notification histories, delivery and evaluation-error visibility, deadman alerts for missing data, delayed AWS metric evaluation, and improved filtering and saved views. Dashboard capabilities now include agent-assisted creation, treemaps, cross-source variable mapping, richer formatting and threshold controls, and easier addition of exploration queries, while new retention controls, storage tiers, and ingestion analysis aim to improve cost and data-management oversight. AWS Cost is introduced as a first-class data source for querying spend alongside telemetry, monitoring budget breaches, and relating cost changes to workload activity, while a managed Dashboard Catalog provides customizable prebuilt dashboards across infrastructure, cloud, database, messaging, CI/CD, AI, and platform domains that also give the agent useful operational context.

groundcover has introduced first-class OAuth connectors for Slack, Linear, Notion, Cursor Cloud Agents, and Pylon, enabling its observability agent to combine telemetry with organizational context such as service ownership, internal documentation, prior decisions, and incident discussions. Administrators can make connectors available organization-wide while controlling individual actions through permissions, approval gates, and blocks, particularly for write operations such as creating tickets. In Slack, users can ask the agent to investigate alerts, retrieve logs, traces, RUM sessions, and metrics, create visualizations or dashboards, and coordinate findings directly in incident threads; it can then draft Linear tickets containing root-cause analysis, evidence, proposed code changes, and follow-up actions. The post presents an example in which the agent identified a checkout outage caused by an active outage-simulation flag in a payments service, traced its cascading effects, and created a detailed ticket after explicit approval. It also describes reusable “skills” that automate end-to-end incident workflows across monitoring, Slack, GitHub, and Linear while retaining shared agent memory across the groundcover interface and Slack, positioning the integrations as a way to improve observability and complement rather than replace coding agents.

Groundcover argues that AI agents are increasing rather than replacing the importance of dashboards, because visualizations provide shared, durable context for investigations, communication, and human feedback. Its Slack-integrated agent can investigate alerts using telemetry such as logs, traces, RUM data, and metrics; create and share dashboards and widgets; incorporate user feedback; and perform approved follow-up actions under RBAC controls, while additional connectors bring in code and organizational context from tools including GitHub, Notion, Cursor, and Pylon. The company describes dashboards as a way to unify inconsistent data naming across infrastructure sources through ingest-time normalization and query-time custom mappings, allowing reusable variables such as a single cluster field across migrated or heterogeneous environments. A catalog of prebuilt dashboards for areas including Kubernetes, AWS, databases, CI/CD, and AI workloads is presented as encoded operational knowledge that agents can use to connect signals, such as rising ingestion volume and storage costs, into actionable findings. Recent dashboard enhancements, including agent-assisted query-to-dashboard creation, threshold coloring, sortable legends, precision controls, and labeled axes, are intended to improve readability and customization. The post concludes that agents should handle much of the work of creating and analyzing visualizations, while dashboards remain central to collaborative observability and human decision-making.

Bring Your Own Cloud (BYOC) in observability is described as an architecture in which a vendor manages components of its platform inside a customer’s cloud environment while the customer retains control of data residency, rather than either fully self-hosting the system or using a vendor’s SaaS query infrastructure. The post distinguishes “true” BYOC from “bring your own bucket” models, arguing that storing raw data in a customer-owned bucket is insufficient if queries and processing still occur in the vendor’s cloud. It identifies managed operations, non-volume-based economics enabled by customer-hosted compute, and data residency as key benefits, and notes that AI-based observability features should run within the customer environment to avoid exporting telemetry, prompts, and related proprietary context. It also outlines a scoped sub-account approach in which vendor access is isolated and controlled through cloud-provider identity and permission tools without extending directly to the data plane. Organizations evaluating BYOC are encouraged to clarify operational responsibilities, data-plane and query locations, pricing, signal coverage, instrumentation requirements, and where AI models run.

Groundcover Skills are reusable instruction sets designed to prevent on-call engineers from repeatedly reconstructing incident-investigation procedures, allowing the groundcover Agent to apply consistent triage steps, conventions, and output formats when relevant. The post situates this approach alongside similar Skills features from Anthropic and OpenAI, describing Skills as portable, task-specific workflows distinct from broader Custom GPT-style assistants. It recommends precise activation triggers, clearly bounded scope, imperative sequential instructions, fixed output templates, explicit defaults and exceptions, narrowly focused workflows, and iterative refinement to improve reliability. In groundcover, Skills are currently private to individual users and include a name, activation criteria, optional description, and Markdown instructions; they can be invoked manually or selected automatically based on a prompt. Examples include a cloud-resource footprint report and Kafka consumer-lag triage, while the company anticipates future shared organizational Skill repositories, monitor-linked investigation recommendations, and integrations that connect investigations to code fixes.