platformchangelog.dev
43 tools

Platform Changelog: what shipped in platform engineering, developer experience and DevOps

feedbefore 2026-10-20all releases, patches included
newest entries
2026-10-1912 days ago
Amazon CloudWatch Omni is now available in additional AWS RegionsannouncementAWS says Amazon CloudWatch Omni, an AI-powered observability experience, is now available in eight more AWS Regions. The expansion adds Canada (Central), Asia Pacific (Mumbai, Singapore, Sydney, Tokyo), Europe (Frankfurt, London) and South America (São Paulo), on top of US East (N. Virginia), US West (Oregon) and Europe (Ireland). The source also describes spaces for cross-account telemetry, agent observability and an IDE extension, and says it works with Azure and GCP telemetry.platform teamsPlatform teams in the newly added Regions can now use Omni for cross-account and multi-cloud observability and AI-assisted root cause analysis. Teams should check pricing and data residency needs before adopting it, since the source only points to the pricing page.What's New with AWS ↗AWS DevOps Agentwritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
InfoQ: Buildkite Adds Scoped Secrets and New Agentic CI ToolspostBuildkite has introduced pipeline-scoped secrets and new Model Context Protocol tools for validating pipelines, diagnosing failed builds and repairing webhooks. It also added Linux arm64 support to its GitHub Actions compatibility layer and expanded OpenTelemetry data in Buildkite Agent v4.1.0.platform teamsPlatform teams running Buildkite can scope secrets to individual pipelines and let MCP-connected agents validate pipelines and diagnose failures. Teams relying on the GitHub Actions compatibility layer on arm64 runners and on Buildkite Agent telemetry can now evaluate those changes.InfoQ ↗written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
2026-10-1615 days ago
Using Claude Code for Salesforce? Don’t miss these security and governance gapspostGearset's blog post lists governance and security gaps for teams using Claude Code, Cursor or Codex to build on Salesforce. The gaps include over-broad local permissions, prompt injection, changes that bypass testing and review, inconsistent team standards, and unplanned maintenance. The post recommends running agents in a container, using least-privilege integration users, restricting MCP toolsets, and routing all AI-generated changes through a CI/CD pipeline.platform teamsPlatform teams that let agents reach Salesforce orgs through MCP should scope credentials to least privilege, restrict enabled MCP toolsets, and require AI-generated changes to pass the same pipeline checks as human changes. The post's recommendations are general, and it promotes Gearset's own product as the solution.Gearset Blog ↗GearsetSalesforce PlatformSalesforce DX MCP ServerAgentforcewritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
2026-10-1516 days ago
GitLab19.5minorGitLab 19.5 was released on October 15, 2026. The GitLab MCP server and GitLab Secrets Manager are now generally available, and the MCP server gives external agents such as Claude Code and Cursor access to a wider set of GitLab tools. The release also adds beta support for connecting externally connected agents to AI Governance, plus non-default branch vulnerability tracking and several GitLab Duo CLI features.platform teamsPlatform teams can scope CI secrets per job by environment and branch and get audit logs for each secret operation, using existing GitLab permissions. Security teams can see sessions from agents running outside GitLab, with audit events sent to a SIEM, though this ECA capability is in beta.written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
CNCF: Extending Kubernetes policy enforcement to runtime security for AI workloadspostNirmata published an open source project, Nirmata Runtime for Kyverno, that uses eBPF and Linux Security Modules to enforce policy on process execution, file access, network egress, and application protocol inside running pods. Policies use the same CEL-based syntax as Kyverno admission policies and start in monitor mode, with denials attributed to the responsible policy and reported as OpenReports objects. The post describes example policies for AI agent pods, such as limiting egress to an LLM gateway, restricting launchable binaries, and detecting agent CLIs in images.platform teamsPlatform teams running agent pods could use per-pod runtime policies to restrict egress, executables, and credential file access beyond what admission control checks. The post provides no benchmarks, adoption data, or independent validation, so the enforcement claims are unverified.CNCF Blog ↗Kyvernowritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
Launch AWS DevOps Agent to review findings and test results from AWS Resilience HubannouncementAWS Resilience Hub now integrates with AWS DevOps Agent. Users can launch a DevOps Agent investigation when they run a resilience test, which returns a structured analysis of symptoms, root cause, impact and observed recovery actions such as ARC Region switch. Users can also start an on-demand DevOps Agent task from a failure mode assessment finding to ask questions and build a mitigation plan.platform teamsPlatform and SRE teams using Resilience Hub on AWS can get agent-driven root cause analysis for resilience tests and help turning failure mode findings into mitigation plans. The source gives no details on permissions, audit or cost of the agent.What's New with AWS ↗AWS DevOps AgentAmazon EKSwritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
Get every merge request in front of the right reviewer, fasterpostGitLab 19.5 makes Recommend Reviewers Flow generally available on GitLab.com, Self-Managed and Dedicated. When a merge request is created, the flow suggests one reviewer based on workload, time zone, availability and expertise defined by approval rules and CODEOWNERS, and shows its reasoning. Authors can assign the suggested reviewer with one click. The vendor also states that Code Review Flow costs a flat 0.25 GitLab Credits per review and cuts per-review cost by 99% against a manual review estimate.platform teamsPlatform teams using GitLab can enable the flow as a project trigger and rely on CODEOWNERS and approval rules as the input for reviewer routing. The benefit to review load and time to first review is the vendor's claim and no independent data is given.GitLab Blog ↗GitLabeditorCorrection 2026-10-20: this was listed as a GitLab 19.5 release by mistake. It is GitLab Blog coverage, not release notes from the project.written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
Unblock agentic software delivery with governed code reviewannouncementGitLab says Code Review Flow, part of GitLab Duo Agent Platform, is now generally available and runs as a CI/CD job on GitLab-hosted or self-hosted runners. Teams can set custom review instructions at the project, group or instance level, and choose GitLab-managed or self-hosted models. The post lists roadmap items including extended analysis, approve or request-changes decisions, risk assessment and local review in Duo CLI.platform teamsPlatform teams can run automated agentic review inside existing merge request, pipeline and approval controls, and can select self-hosted models for air-gapped or regulated setups. The $0.25 per review price and 100x ROI figure are vendor claims, and the source gives no methodology for them.GitLab Blog ↗GitLabGitHub Actionswritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
Connect your agents to your software lifecycle with the GitLab MCP serverpostGitLab announced general availability of its MCP server with GitLab 19.5, letting MCP-compatible agents work with code, merge requests, work items and pipelines. The release adds tools for branches, file reads, semantic code search, pipeline inspection and artifact reads, and consolidates related tools into fewer calls. Agents can also start and follow GitLab Duo Agent Platform flows, with actions running under existing GitLab permissions.platform teamsPlatform teams evaluating agent access to GitLab can connect external agents with permissions inherited from GitLab, and can limit the tool surface with toolsets. The source gives no measured figures for the claimed reduction in calls or context.GitLab Blog ↗GitLabeditorCorrection 2026-10-20: this was listed as a GitHub Actions, GitLab 19.5 release by mistake. It is GitLab Blog coverage, not release notes from the project.written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
2026-10-1318 days ago
Amazon Bedrock AgentCore Memory now supports configurable time-to-live for long-term memoryannouncementAmazon Bedrock AgentCore Memory now supports a configurable time-to-live on long-term memory records, so records are deleted automatically after a retention period the user sets. Previously long-term memory records did not expire by default. TTL can be set on the resource or on specific records, and changing it requires the new bedrock-agentcore:UpdateMemoryRecordExpirySpecification permission. The feature is available in all AWS Regions where AgentCore Memory is supported.platform teamsPlatform teams can set retention rules to meet data-retention requirements and limit storage growth from agent memory. Because changing TTL needs its own IAM permission, access to retention settings can be granted separately from general memory access.What's New with AWS ↗written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
GitLab's agent orchestration eliminates the wait between development stagesannouncementGitLab announced agent orchestration capabilities in GitLab Duo Agent Platform, including goal-driven flows that run review, tests, security scans and approvals as one sequence. It also announced GitLab for Slack, which starts flows from Slack threads, and Custom Flows, which can be triggered by mentions, assignments or pipelines. The source says Custom Flows are generally available, while the MCP server and GitLab for Slack are slated for general availability later this month.platform teamsPlatform teams running GitLab can evaluate agents that act on pipeline failures and merge workflows, with the stated requirement of GitLab 19.3 or later and Duo CLI 9.17.0 or later on Premium or Ultimate. The source gives no metrics for the claimed delivery speedups, so teams would need their own measurements before relying on them.GitLab Blog ↗GitLabwritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
Make every AI credit count: See, cap, and reduce agent spendannouncementGitLab describes new controls in its Duo Agent Platform for managing agent spend. These include per-user and per-group credit caps with alerts at 50%, 80% and 100% and an automatic pause at the cap, plus Impact Analytics for breaking spend down by team, task and model. GitLab also says three open-weight models are now hosted, with internal benchmark claims of matching frontier models at lower cost, and that its Orbit context graph reduces agent retries.platform teamsPlatform teams using GitLab can pin models per feature and set per-user credit caps today, with Impact Analytics in early access. Orbit is in free beta on GitLab.com and is slated for general availability next month, and its cost benefits rest on GitLab's own testing.GitLab Blog ↗GitLabwritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
Scaling Git for the agentic software factoryannouncementGitLab describes next-gen SCM, a redesigned Git backend built to serve thousands of concurrent agents while keeping the Git protocol, and says it is in private beta. The post cites Anthropic's monorepo growth from about 2,500 daily merges in May to more than 30,000 in September, and Anthropic's own Scion system built for that monorepo. GitLab says early internal tests showed up to 50x faster workloads, up to 2x fewer tokens and up to 1,000x less network traffic, described as peak results under specific conditions.platform teamsPlatform teams evaluating GitLab for agent-heavy workloads can request private beta access to test next-gen SCM against their own repositories. The performance figures are vendor test results and have not been validated under sustained concurrent load, so they should not be used for capacity planning yet.GitLab Blog ↗GitLabwritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
Introducing remote MCP for Octopus DeploypostOctopus Deploy now includes a built-in remote MCP server, so agents can connect over HTTP to an instance's /mcp endpoint without a local install. The server initially exposes a selected set of core features covering spaces, projects, tenants, deployments, runbooks, task logs, audit events and tenant variables, with access depending on the authentication method and granted permissions. Octopus Cloud customers can use it now and Octopus Server customers will get it in the 2026.3 release. The company says scoped API keys, OAuth for the MCP server, agent-aware policies and an improved audit log are planned for the coming months.platform teamsPlatform teams running Octopus can connect agents without distributing and updating a local MCP server on every machine. Governance controls such as scoped API keys, agent-aware policies and agent audit filtering are not yet available, so teams should plan access controls before letting agents trigger deployments.Octopus Deploy Blog ↗Octopus DeployeditorCorrection 2026-10-20: this was listed as a Octopus Deploy release by mistake. It is Octopus Deploy Blog coverage, not release notes from the project.written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
2026-10-1219 days ago
Announcing consumption-based pricing for short-term memory in Amazon Bedrock AgentCoreannouncementAmazon Bedrock AgentCore now bills short-term memory by data ingested, retrieved, and stored, replacing the previous flat rate per event. Ingestion is priced at $1.00 per GB, retrieval at $0.20 per GB, and storage at $0.10 per GB-month, with a 12 KB minimum and 64 KB maximum billable size per event for ingestion and retrieval. AWS says most workloads see lower costs, and long-term memory pricing is unchanged.platform teamsTeams running agents on AgentCore memory need to re-estimate costs, since the billing unit changed from events to data volume and small-payload workloads are the ones AWS says benefit most. Setting a shorter TTL on memory resources is the stated lever for controlling storage spend.What's New with AWS ↗written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
InfoQ: Spotify More Than Doubled Monthly Code Changes as AI Accelerated Developmentcase studySpotify reports that AI-assisted development more than doubled its monthly merged changes. The company says it is reviewing quality and reliability controls across its engineering organization in response. Described changes include testing, observability, rollback capacity, workload prioritization and automated maintenance.platform teamsPlatform teams facing a similar rise in change volume can compare their testing, observability and rollback practices against the areas Spotify names. The source gives no specific tooling, metrics or figures beyond the merged-change increase.InfoQ ↗written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
CNCF: From zero to agent builder: Running kagent on your laptoppostThe author walks through running kagent, a CNCF Sandbox project, on a local kind cluster with a local Ollama model. An agent-builder agent is defined as a Kubernetes Agent resource that can only apply Agent manifests, with a three-tool allowlist and human approval before each apply. The post notes that security settings such as default unsecure auth and a built-in PostgreSQL password are deliberately left out of this laptop setup.platform teamsA platform team evaluating agents-as-resources can see how RBAC limited to agents.kagent.dev, a tool allowlist and requireApproval gate constrain what a generated agent can do. The post states the laptop setup is not suitable for a shared cluster without the security settings it omits.CNCF Blog ↗kagentKubernetes MCP Serverwritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
InfoQ: GitLab 19.4 Adds Goal-Driven Agents and Tighter Cost ControlspostGitLab released version 19.4 of its DevSecOps platform. The release adds a goal-driven mode to GitLab Duo CLI, three hosted open-weight models, a larger set of Model Context Protocol tools, and more detailed controls for AI spending.platform teamsPlatform teams running GitLab can now set AI spending controls and use a broader MCP tool set when agents consume GitLab APIs. The source text does not give details on the controls or tool permissions, so the specifics need checking in the release notes.InfoQ ↗GitLabeditorCorrection 2026-10-20: this was listed as a GitLab 19.4 release by mistake. It is InfoQ coverage, not release notes from the project.written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
2026-10-1120 days ago
Agentic AI in DevOps: Exploring Use Cases for the CloudBees MCPpostCloudBees published a blog post describing use cases for its Unify platform combined with an MCP server, focused on pipeline diagnostics and security remediation. The source text is a short teaser and does not describe specific features, releases, or results. The vendor says the combination reduces DevOps tool sprawl and keeps developers in flow without context switching, but these are unverified claims.platform teamsA platform team evaluating CloudBees could look at whether an MCP server for pipeline diagnostics fits its tooling, but the source gives no concrete capabilities, permission model, or governance details to assess. No concrete change can be confirmed from this text.CloudBees Blog ↗CloudBeeswritten by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
InfoQ: Foresight AI Brings Gremlin Agents to Reliability EngineeringpostGremlin has announced general availability of Foresight AI, an agentic product. According to the source, it analyzes services for potential failures, recommends changes, and reruns tests to check whether a fix works.platform teamsPlatform teams evaluating AI-driven reliability tooling can now consider Foresight AI as a GA option, but the source gives no details on integrations, pricing, or how it connects to existing pipelines or clusters.InfoQ ↗written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
2026-10-0823 days ago
CNCF: Don’t give AI agents root. Make them propose the next system state.postThe author argues that autonomous agents should not be given root access to production machines, since agent behavior is nondeterministic and root access offers no reproducibility. He proposes that agents run in short-lived isolated workloads, observe immutable hosts, and submit pull requests for the next system image, which then passes review, build, signing and release before the host consumes it. He states that the agent identity should be able to propose changes without approving, merging, altering pipelines, or signing images. The post is written by a Kairos maintainer and describes a go-ukify alignment fix found by an agent during QA.platform teamsPlatform teams evaluating agents on infrastructure can treat agent write access as a separation-of-authority question: agents propose changes through the same review and build pipeline as humans, with no direct host mutation. The post describes an architectural pattern rather than a shipped product feature.CNCF Blog ↗written by claude-haiku-5-5, prompt 7cbe154, checked by publish rules
daily email

One short email each morning: what shipped, what breaks, and a link to each change. Unsubscribe in one click. Privacy.

statuslast 30 days for lines
tools tracked43releases in registry964released in the last 30 days106entries published336breaking change lines18security lines84
change typestwo-column gutter, like git status
  • +addedNew functionality.
  • ~changedExisting behavior changed.
  • -deprecatedStill works, scheduled for removal.
  • -removedDecommissioned, gone in this version.
  • *fixedA defect corrected.
  • !securityA vulnerability closed or a dependency patched for one.
  • !breakingFlag in the first column; the change needs action before upgrading.
filter
categoryallidp & portalsai agents on platformci/cdobservabilitysupply chaincost & governancereports & dx dataenterprise app platforms
showmajor onlyreleasescase studiespostmortemsinclude patches