# Everruns Cloud changelog

> One entry per finished change, newest first. Each line: date, products, kind, and where the change is available.

Source: <https://everruns.com/docs/changelog/cloud/>

Feed: <https://everruns.com/docs/changelog/cloud/rss.xml>

- [Session Trace names prompt cache resets and their cause](https://everruns.com/docs/changelog/2026-10-12-session-trace-cache-resets/): 2026-10-12 · Platform, Cloud · Improvement. When a model call loses the prompt cache the previous call built, the trace now says so and names the likely cause, such as changed tools or an idle gap.
- [Prompt caching on Bedrock, OpenRouter Claude and Mistral](https://everruns.com/docs/changelog/2026-10-12-prompt-caching-bedrock-openrouter-mistral/): 2026-10-12 · Platform, Cloud, Framework · Improvement. Prompt caching now writes cache points on Bedrock Claude and Nova, turns on caching for Claude through OpenRouter, and routes Mistral by cache key.
- [Prompt caching gets a retention setting and a prompt-only mode](https://everruns.com/docs/changelog/2026-10-12-prompt-cache-retention/): 2026-10-12 · Platform, Cloud, Framework · Improvement. Ask for one hour Anthropic cache entries with retention 1h, and use the explicit strategy on Claude to cache tools and instructions but not the conversation.
- [GitHub triggers narrow CI events by branch and workflow](https://everruns.com/docs/changelog/2026-10-12-github-trigger-ci-filters/): 2026-10-12 · Platform, Cloud · Improvement. The GitHub trigger form takes the branches and workflow names a CI trigger accepts, offers one session for every event, and the agent builder fills both in.
- [GitHub tools act as one chosen identity, with narrowed tokens](https://everruns.com/docs/changelog/2026-10-12-github-tools-act-as/): 2026-10-12 · Platform, Cloud · Improvement. A new acts_as setting picks whose GitHub identity the tools use, and each call gets a token limited to one repository and the permissions it needs.
- [Docs moved to everruns.com/docs](https://everruns.com/docs/changelog/2026-10-12-docs-move-to-everruns-com/): 2026-10-12 · Cloud, Platform, Framework · Improvement. The documentation now lives at everruns.com/docs, and every docs.everruns.com link, Markdown file and schema redirects to the same page there.
- [Sending a message right after cancelling a turn no longer hangs](https://everruns.com/docs/changelog/2026-10-12-cancelled-turns-no-longer-strand/): 2026-10-12 · Platform, Cloud · Fix. A message sent just after a cancel now always starts its own turn, and a cancelled turn stops writing output once it reports the cancel.
- [Browser agents see their screenshots and keep their page](https://everruns.com/docs/changelog/2026-10-12-browser-agents-screenshots-focus/): 2026-10-12 · Platform, Cloud · Fix. Browser screenshots now reach the model, Browserless tools no longer race on one page, and key presses report where focus landed.
- [Choose where an agent's GitHub App lives, and manage it from the CLI](https://everruns.com/docs/changelog/2026-10-12-agent-github-org-cli/): 2026-10-12 · Platform, Cloud, CLI · Improvement. Connect GitHub suggests the organizations your team already uses for an agent's App, an uninstalled App can start over, and the CLI manages it.
- [Load Everruns workflow skills through MCP](https://everruns.com/docs/changelog/2026-10-11-mcp-workflow-skills/): 2026-10-11 · Platform, Cloud · Feature. The Everruns MCP endpoint serves two workflow skills for running sessions and authoring agents, with metadata and verified file manifests.
- [Call one agent from code with an Agent API key](https://everruns.com/docs/changelog/2026-10-11-agent-api-channel/): 2026-10-11 · Platform, Cloud · Feature · Major. An Agent API channel gives an application a key that reaches one agent's sessions, with streaming, approvals, end users, browser origins and spend limits.
- [Tools in Shell runs an agent's tools from Bashkit scripts](https://everruns.com/docs/changelog/2026-10-10-tools-in-shell/): 2026-10-10 · Platform, Cloud · Feature · Major. The tools_in_shell capability moves most tools into one tools shell command, so a script can chain calls, preview them with tools plan and ask approval first.
- [Session Trace view replaces Transcript and Timeline](https://everruns.com/docs/changelog/2026-10-10-session-trace-view/): 2026-10-10 · Platform, Cloud, CLI · Feature · Major. Sessions open on a Trace tab that lays out turns and steps with timing bars, an inspector for each step's input, output and model request, and event search.
- [Session SQL databases survive server restarts](https://everruns.com/docs/changelog/2026-10-10-session-sql-databases-persist/): 2026-10-10 · Platform, Cloud · Fix. Session SQL databases are now stored in PostgreSQL page by page, so a server restart no longer loses them and every server instance sees the same data.
- [Everruns agents act as Poppy personal agents](https://everruns.com/docs/changelog/2026-10-10-poppy-personal-agents/): 2026-10-10 · Platform, Cloud · Feature. The poppy_client capability lets an Everruns agent find a company's agent through the Personal Agent Protocol, talk to it and hand off to a person there.
- [Add your own MCP servers and use them in chat](https://everruns.com/docs/changelog/2026-10-10-my-mcp-servers/): 2026-10-10 · Platform, Cloud · Feature · Major. People add personal MCP servers in Settings, and agents with the User MCP Servers capability use them, including servers connected in the middle of a chat.
- [Label MCP tools as read only or changes things](https://everruns.com/docs/changelog/2026-10-10-mcp-tool-risk-labels/): 2026-10-10 · Platform, Cloud, CLI · Feature. Save a risk label on each tool of an organization MCP server to decide whether it asks for approval, with model-suggested labels a person can accept.
- [Built-in harnesses reorganized as a tree by use case](https://everruns.com/docs/changelog/2026-10-10-harness-tree/): 2026-10-10 · Platform, Cloud · Improvement. Base, Conversation, Worker, Bashkit Worker and Sandbox Worker now form a tree of built-in harnesses, and existing Worker bindings move by sandbox policy.
- [Pick and configure guardrail checks in the agent editor](https://everruns.com/docs/changelog/2026-10-10-guardrail-checks-editor/): 2026-10-10 · Platform, Cloud · Improvement. The Guardrails settings panel lets authors add presets or single checks and set their stage and action, and model-backed checks can now judge the final reply.
- [GitHub triggers wake agents on CI results](https://everruns.com/docs/changelog/2026-10-10-github-ci-triggers/): 2026-10-10 · Platform, Cloud · Feature. GitHub triggers now receive workflow run and check suite completions, and a changes-only filter wakes the agent only when a workflow's CI result changes.
- [Explicit communication lets agents choose when to reply](https://everruns.com/docs/changelog/2026-10-10-explicit-communication/): 2026-10-10 · Platform, Cloud, Framework, CLI · Feature. An agent's Communication setting can be Explicit, so assistant text stays as private notes and the agent speaks only through send_message or stays silent.
- [Outbound network policy modes, audit log and org allowlists](https://everruns.com/docs/changelog/2026-10-10-egress-policy-controls/): 2026-10-10 · Platform, Cloud, CLI · Feature. Operators choose open or curated outbound access, every outbound request is audited with its org and session, and admins can widen the allowlist for one org.
- [Everruns Cloud verify-email screen sends a new link again](https://everruns.com/docs/changelog/2026-10-10-cloud-verify-email-resend-fix/): 2026-10-10 · Cloud · Fix. The email verification step on Everruns Cloud now offers Sign in again, which sends a fresh link, replacing a resend action that never sent any mail.
- [Everruns Cloud shows spend per model and agent on a Usage tab](https://everruns.com/docs/changelog/2026-10-10-cloud-usage-tab-credit/): 2026-10-10 · Cloud · Feature. A Usage tab breaks down Everruns Cloud charges by day, model and agent with token counts and CSV export, and a credit chip and banner warn on low credit.
- [ChatGPT plan connections no longer need a feature flag](https://everruns.com/docs/changelog/2026-10-10-chatgpt-plan-no-flag/): 2026-10-10 · Platform, Cloud · Improvement. The ChatGPT provider now shows in Connect provider on every deployment, including Everruns Cloud, because signing in is already personal to each user.
- [Rate agent replies and branch a chat from any message](https://everruns.com/docs/changelog/2026-10-10-chat-feedback-and-branching/): 2026-10-10 · Platform, Cloud, CLI · Improvement. Agent replies in Chat get Good, Bad and Branch into new chat actions, and each message shows one status row from sending until the answer arrives.
- [api_endpoint channels convert to Agent API channels](https://everruns.com/docs/changelog/2026-10-10-api-endpoint-channels-converted/): 2026-10-10 · Platform, Cloud · Breaking. Existing api_endpoint channels become Agent API channels at startup with the same keys, and old request shapes now return api_endpoint_retired or HTTP 410.
- [Agents must give a reason for every change by default](https://everruns.com/docs/changelog/2026-10-10-agent-change-reasons-required/): 2026-10-10 · Platform, Cloud · Breaking. Require reasons from agents is now on by default, so an agent's change without a reason fails with reason_required; organizations can turn the flag off.
- [Agent budgets stop turns before each model call](https://everruns.com/docs/changelog/2026-10-10-agent-budgets-enforced/): 2026-10-10 · Platform, Cloud, Framework · Fix. Agent budgets are now checked before every model call in the native agent loop, so an exhausted budget stops the turn with a typed reason and budget events.
- [Voice channels let people call an agent](https://everruns.com/docs/changelog/2026-10-09-voice-channels/): 2026-10-09 · Platform, Cloud, Framework · Feature · Major · In 0.47.0. A voice channel lets people talk to an agent over WebRTC while it keeps its own model, tools and instructions, on the platform and in the Framework.
- [Models and providers move into one registry page](https://everruns.com/docs/changelog/2026-10-09-models-providers-registry/): 2026-10-09 · Platform, Cloud · Improvement · In 0.47.0. Registries → Models now holds Models, Providers and Defaults tabs, so connecting a provider, choosing its models and setting org defaults happen in one place.
- [See and revoke AI clients connected to Everruns](https://everruns.com/docs/changelog/2026-10-09-connected-ai-clients/): 2026-10-09 · Platform, Cloud · Feature · In 0.47.0. Settings → My agent experience lists the AI clients you connected to Everruns over MCP OAuth, shows when each was last used, and lets you revoke one.
- [app.everruns.com serves llms.txt and auth.md for AI agents](https://everruns.com/docs/changelog/2026-10-09-cloud-agent-discovery-routes/): 2026-10-09 · Cloud · Improvement. AI agents and crawlers can discover Everruns Cloud from app.everruns.com itself through llms.txt, auth.md, Link headers, robots.txt and a sitemap.
- [Fix ChatGPT plan sessions failing with HTTP 400](https://everruns.com/docs/changelog/2026-10-09-chatgpt-plan-http-400-fix/): 2026-10-09 · Platform, Cloud, Framework · Fix · In 0.47.0. Sessions that use a ChatGPT plan connection no longer fail with HTTP 400, because Everruns stops sending request-body metadata the Codex backend rejects.
- [Agents home and a New agent page with an agent builder](https://everruns.com/docs/changelog/2026-10-09-agents-home-new-agent/): 2026-10-09 · Platform, Cloud, CLI · Feature · Major · In 0.47.0. The Agents page now shows run activity, channels and setup problems for each agent, and New agent drafts an agent from a description, an example or an import.
- [Save shell scripts on agents and run them from triggers](https://everruns.com/docs/changelog/2026-10-09-agent-saved-scripts/): 2026-10-09 · Platform, Cloud, CLI · Feature · In 0.46.0. Agents keep named shell scripts that run in Bashkit with their tools, and schedule or webhook triggers can run one with the event payload and no model call.
- [Desktop computer use on E2B and Daytona sandboxes](https://everruns.com/docs/changelog/2026-10-08-desktop-computer-use/): 2026-10-08 · Platform, Cloud · Feature. Experimental capabilities run the computer tool on a full Linux desktop in an E2B or Daytona sandbox, now with zoom, cursor, mouse button and hold-key actions.
- [Coordinator threads in Platform Chat and the Framework](https://everruns.com/docs/changelog/2026-10-08-coordinator-threads/): 2026-10-08 · Platform, Cloud, Framework · Feature · In 0.45.0. Platform Chat and Framework agents can hand focused work to child threads and keep a board of what is working, what needs you and what is ready for review.
- [Browser tool for agents on Browserless](https://everruns.com/docs/changelog/2026-10-08-browser-use-capability/): 2026-10-08 · Platform, Cloud · Feature · In 0.45.0. The experimental browser_use capability lets an agent read pages as an accessibility tree, act on elements by ref, fill forms and manage tabs in Browserless.
- [Claude Haiku 5.5 model added for Anthropic providers](https://everruns.com/docs/changelog/2026-10-07-claude-haiku-5-5/): 2026-10-07 · Platform, Cloud, Framework · Feature · In 0.44.0. The claude-haiku-5-5 model and its 1M-context twin are seeded as an enabled favorite, with adaptive thinking, tool search and the computer toolset.
- [Sign in to channels and Public Chat with AgentID](https://everruns.com/docs/changelog/2026-10-07-agentid-sign-in/): 2026-10-07 · Platform, Cloud · Feature · In 0.44.0. A2A, AG-UI and Public Chat channels accept AgentID tokens, Public Chat shows Continue with AgentID, and agents can sign in to apps with their AgentMail inbox.
- [A2A channels serve HTTP+JSON and the PACT profiles](https://everruns.com/docs/changelog/2026-10-07-a2a-http-json-and-pact/): 2026-10-07 · Platform, Cloud · Feature · In 0.44.0. Every A2A channel also answers the A2A 1.0 HTTP+JSON binding, and a channel can accept personal agents through the PACT Identity and Delegated profiles.
- [Change history with reasons and restore for every entity](https://everruns.com/docs/changelog/2026-10-06-change-history-with-reasons/): 2026-10-06 · Platform, Cloud, CLI · Feature · Major · In 0.42.0. Every configuration change records who made it, through which surface and why, with numbered revisions you can compare and restore from the History menu.
- [Agent versions are retired in favor of change history](https://everruns.com/docs/changelog/2026-10-06-agent-versions-retired/): 2026-10-06 · Platform, Cloud, CLI · Breaking · In 0.44.0. Agent publishing, default versions and version pins are removed; channels, triggers and sessions now run the agent's current configuration instead.
- [Sandbox Templates and a Sandboxes fleet view](https://everruns.com/docs/changelog/2026-10-05-sandbox-templates-fleet-view/): 2026-10-05 · Platform, Cloud · Feature · Major · In 0.42.0. Reusable Sandbox Templates with immutable revisions decide where an agent's commands run, and a fleet page lists every sandbox in the organization.
- [Agents get avatars from presets or an uploaded image](https://everruns.com/docs/changelog/2026-10-05-agent-avatars/): 2026-10-05 · Platform, Cloud · Feature · In 0.42.0. Give an agent an avatar from 25 curated presets or a cropped upload, and it shows in the console, on the A2A Agent Card and as the Slack app icon.
- [Slack permission issues appear in Health with recovery steps](https://everruns.com/docs/changelog/2026-10-03-slack-health-issues-recovery/): 2026-10-03 · Platform, Cloud · Feature · In 0.39.0. Slack installations missing a permission or credential raise a persistent issue in Settings > Health, with guided recovery, per-user snoozing and rechecks.
- [Playground for testing agents, and Chat as one permanent conversation](https://everruns.com/docs/changelog/2026-10-03-playground-agent-testing/): 2026-10-03 · Platform, Cloud · Feature · Major · In 0.38.0. Agent testing moves to a shared Playground where members pick the agent and the user it runs as, while Chat becomes your permanent Platform Chat conversation.
- [Conversation becomes the default harness and Generic is deprecated](https://everruns.com/docs/changelog/2026-10-03-conversation-default-harness/): 2026-10-03 · Platform, Cloud, Framework · Breaking · In 0.38.0. Built-in harnesses form progressive presets from Base to Worker, Conversation is the new default, and agents that inherited Generic are pinned to it on upgrade.
- [Computer use capability lets agents operate a browser screen](https://everruns.com/docs/changelog/2026-10-03-computer-use-capability/): 2026-10-03 · Platform, Cloud · Feature · In 0.35.0. The experimental computer_use capability drives a Browserless page with clicks, typing and screenshots, using native OpenAI or Anthropic tools when available.
- [Chat threads keep conversations and ongoing work beside Chat](https://everruns.com/docs/changelog/2026-10-03-chat-threads-workspace/): 2026-10-03 · Platform, Cloud · Feature · In 0.39.0. Organizations can opt into Chat threads, a panel beside the permanent Chat for searching conversations, following background work and resolving threads.
- [Approval history in chat and a new Approvals audit page](https://everruns.com/docs/changelog/2026-10-03-approval-history-audit-page/): 2026-10-03 · Platform, Cloud · Feature · In 0.37.0. Chats show approval requests and recorded consent as clear entries, and organization admins get an Approvals page that filters approvals across sessions.
- [Cloudflare AI Gateway and Vercel AI Gateway model providers](https://everruns.com/docs/changelog/2026-10-02-cloudflare-vercel-ai-gateway/): 2026-10-02 · Platform, Cloud, Framework · Feature · In 0.34.2. Agents can run models through Cloudflare AI Gateway or Vercel AI Gateway, and the Cloudflare provider lists Workers AI text-generation models automatically.
- [AG-UI 1.0 endpoint and AG-UI streaming for Framework sessions](https://everruns.com/docs/changelog/2026-10-02-ag-ui-1-0-protocol/): 2026-10-02 · Platform, Cloud, Framework · Feature · Major · In 0.34.0. The AG-UI channel moves to protocol 1.0 with interrupts, resume, frontend tools and subagents, and any Framework session can serve AG-UI from your own server.
- [A2A 1.0 endpoint with task results, streaming and push notifications](https://everruns.com/docs/changelog/2026-10-02-a2a-1-0-protocol/): 2026-10-02 · Platform, Cloud, Framework · Feature · Major · In 0.34.0. Everruns A2A channels negotiate protocol 1.0, return agent output as task artifacts, support ListTasks and SubscribeToTask, and send push notifications.
- [Connect several Slack workspaces per organization](https://everruns.com/docs/changelog/2026-10-01-slack-workspaces-per-organization/): 2026-10-01 · Platform, Cloud · Feature · In 0.34.0. An organization connects its Slack workspaces once in Settings, and adding an agent to Slack becomes picking a workspace and approving one consent screen.
- [OpenAI structured output and background mode for long reasoning](https://everruns.com/docs/changelog/2026-10-01-openai-structured-output-background/): 2026-10-01 · Platform, Cloud, Framework · Feature · In 0.34.0. OpenAI drivers enforce JSON Schema structured output, and xhigh or max reasoning calls run as background responses that resume after a drop or worker restart.
- [MCP Events add session webhooks and triggers on MCP server events](https://everruns.com/docs/changelog/2026-10-01-mcp-events-webhooks-triggers/): 2026-10-01 · Platform, Cloud · Feature · In 0.34.0. With the experimental MCP Events flag, MCP clients subscribe to signed session webhooks, and agents wake on events published by their own MCP servers.
- [PR reviewer and security scanner agent templates for GitHub](https://everruns.com/docs/changelog/2026-10-01-github-review-security-templates/): 2026-10-01 · Platform, Cloud · Feature · In 0.34.0. Two guided agent templates use the agent's GitHub App to review pull requests with inline comments and to run a weekly scan that files GitHub issues.
- [OpenAI hosted code interpreter, shell, file search and remote MCP](https://everruns.com/docs/changelog/2026-09-30-openai-hosted-tools-remote-mcp/): 2026-09-30 · Platform, Cloud · Feature · In 0.33.0. The openai_server_tools capability adds code interpreter, hosted shell, file search and approval-gated remote MCP next to web search on OpenAI Responses agents.
- [GitHub pull request triggers with a per-agent GitHub App](https://everruns.com/docs/changelog/2026-09-30-github-pull-request-triggers/): 2026-09-30 · Platform, Cloud · Feature · Major · In 0.33.0. Each agent can create and install its own GitHub App by clicks, read pull requests and diffs, post comments, and start sessions from pull request events.
- [Claude Sonnet 5.5 and GPT-6.1 Sol with fast and ultrafast tiers](https://everruns.com/docs/changelog/2026-09-30-claude-sonnet-5-5-gpt-6-1-sol/): 2026-09-30 · Platform, Cloud, Framework · Feature · In 0.33.0. Claude Sonnet 5.5 becomes the favorite Sonnet model, GPT-6.1 Sol adds fast and ultrafast speed tiers, and the sunset Claude Sonnet 4.5 profile is removed.
- [Everruns MCP server renders as an MCP App in ChatGPT and Codex](https://everruns.com/docs/changelog/2026-09-29-mcp-app-chatgpt-codex/): 2026-09-29 · Platform, Cloud · Feature · In 0.33.0. MCP App hosts such as ChatGPT and Codex show a live session view, answer agent questions in a form, approve pending actions and open an Everruns panel.
- [Agent page rebuilt as one view-and-edit workspace](https://everruns.com/docs/changelog/2026-09-29-agent-page-workspace/): 2026-09-29 · Platform, Cloud · Improvement · In 0.33.0. The agent page shows the system prompt as the page with settings in a side column, and you edit everything in place with one Save or Discard.
- [Ask User questions reach MCP, A2A and Tasks clients](https://everruns.com/docs/changelog/2026-09-25-ask-user-mcp-a2a-clients/): 2026-09-25 · Platform, Cloud · Improvement · In 0.32.0. A session waiting on ask_user now shows its question to MCP clients as a form elicitation, to A2A callers as a typed data part, and to Tasks clients.
- [Ask User lets agents pause and ask you questions mid-run](https://everruns.com/docs/changelog/2026-09-24-ask-user-agent-questions/): 2026-09-24 · Platform, Cloud, Framework · Feature · Major · In 0.31.0. Agents with the ask_user capability pause a run to ask one to four choice, text or secret questions, shown as an inline card that you answer in place.
- [Claude always-thinking models no longer return 400s or empty answers](https://everruns.com/docs/changelog/2026-09-23-claude-always-thinking-fixes/): 2026-09-23 · Platform, Cloud, Framework · Fix · In 0.31.0. The Anthropic driver rejects prefill early, maps an explicit none effort to low, and keeps requests append-only on Opus 5.5 and other always-thinking models.
- [Everruns Cloud sign-in failures return to the login page](https://everruns.com/docs/changelog/2026-09-22-cloud-sign-in-failure-recovery/): 2026-09-22 · Cloud · Fix. A failed Everruns Cloud sign-in now lands on the login page with the reason and a way to try again, instead of a raw JSON error on the callback URL.
- [Claude Opus 5.5, GPT-6 Sol and GPT-6 Luna models added](https://everruns.com/docs/changelog/2026-09-22-claude-opus-5-5-gpt-6-models/): 2026-09-22 · Platform, Cloud · Feature · In 0.30.0. Claude Opus 5.5 becomes the recommended Opus model, and GPT-6 Luna replaces GPT-5.6 Terra as the platform default model and the default utility model.
- [Install an Everruns agent in Slack with one click](https://everruns.com/docs/changelog/2026-09-19-slack-one-click-install/): 2026-09-19 · Cloud, Platform · Feature · In 0.30.0. Connecting an agent to Slack no longer needs credentials copied from api.slack.com: Everruns creates the Slack app and takes its credentials from OAuth.
- [MCP catalog presets, acts-as identity and personal connections](https://everruns.com/docs/changelog/2026-09-19-mcp-catalog-connections-acts-as/): 2026-09-19 · Platform, Cloud · Feature · In 0.30.0. MCP servers gain an org catalog of presets, an actsAs setting for whose sign-in a server uses, an agent MCP tab and a place to revoke your own sign-ins.
- [Set spend budgets on agent endpoints from the Integrations tab](https://everruns.com/docs/changelog/2026-09-19-agent-endpoint-budgets/): 2026-09-19 · Platform, Cloud · Feature · In 0.30.0. Agent and endpoint budgets can be managed on the agent's Integrations tab, and sessions are attributed to the endpoint they arrived through.
- [Slack agents ask approval with buttons and show task progress](https://everruns.com/docs/changelog/2026-09-18-slack-approvals-progress-actions/): 2026-09-18 · Platform, Cloud · Feature · In 0.30.0. Agents deployed to Slack can take approvals through Approve and Decline buttons, show live sub-task progress and act as the channel's own bot.
- [Agent endpoints and triggers replace Apps in Everruns](https://everruns.com/docs/changelog/2026-09-18-agent-endpoints-replace-apps/): 2026-09-18 · Platform, Cloud · Breaking · Major · In 0.30.0. App management is retired: an agent's endpoints and triggers now live on its Integrations tab, with an org-wide Exposures view and per-agent suspend.
- [Soft approval: agents pause before destructive or outward actions](https://everruns.com/docs/changelog/2026-09-17-soft-approval-capability/): 2026-09-17 · Platform, Cloud, Framework · Feature · In 0.29.0. The new soft_approval capability, on by default for Generic and Platform Chat, has agents ask before risky actions and records who approved in the audit log.
- [One everruns command line for the CLI, MCP and agent shells](https://everruns.com/docs/changelog/2026-09-17-everruns-command-line-tree/): 2026-09-17 · Platform, Cloud, CLI, Framework · Feature · Major · In 0.29.0. Every Everruns operation now has one noun-verb spelling with --help that works the same in the everruns CLI, the /mcp endpoint and an agent's shell.
- [The Everruns UI supports dark mode with a theme setting](https://everruns.com/docs/changelog/2026-09-16-ui-dark-mode-theme/): 2026-09-16 · Platform, Cloud · Feature · In 0.28.0. The Everruns web UI can now be shown in light, dark or system theme, chosen from the user menu and remembered without a flash on navigation.
- [Everruns agents run in Slack's agent pane](https://everruns.com/docs/changelog/2026-09-15-slack-agent-pane/): 2026-09-15 · Platform, Cloud · Feature · Major · In 0.27.0. A Slack channel can now serve Slack's agent pane, with streamed replies, a stop button, live status and thread context, alongside channel replies.
- [Everruns Cloud adds a built-in model provider with starter credit](https://everruns.com/docs/changelog/2026-09-15-cloud-built-in-model-provider/): 2026-09-15 · Cloud · Feature · Major. New Everruns Cloud organizations run agents without provider keys, paid from prepaid credit that starts at $5 and is topped up in Settings > Billing.
- [Sessions show where their commands run and what they may touch](https://everruns.com/docs/changelog/2026-09-14-session-environment-panel/): 2026-09-14 · Platform, Cloud, Framework · Feature · In 0.27.0. Each session now declares its execution environment, readable through the API and shown in an Environment panel on the session Workspace tab.
- [Everruns Cloud loads Sessions, Harnesses and Skills pages faster](https://everruns.com/docs/changelog/2026-09-14-cloud-faster-page-navigation/): 2026-09-14 · Cloud · Improvement. Moving between the Sessions, Harnesses and Skills pages on Everruns Cloud no longer stalls for seconds while the server renders the page.
- [Send a message and wait for the turn result in one request](https://everruns.com/docs/changelog/2026-09-12-wait-for-message-completion/): 2026-09-12 · Platform, Cloud, Framework · Feature · In 0.26.0. The sessions API can now hold a message request open until the turn finishes, and Framework runs accept a timeout, for callers that do not consume SSE.
- [Attach PDF files to agent messages](https://everruns.com/docs/changelog/2026-09-12-pdf-file-attachments/): 2026-09-12 · Platform, Cloud · Feature · In 0.26.0. Messages can now carry PDF attachments, uploaded through a new files API and passed to each provider in its native document format.
- [Session schedules no longer fire twice on multi-replica deployments](https://everruns.com/docs/changelog/2026-09-11-session-schedules-no-double-fire/): 2026-09-11 · Platform, Cloud · Fix · In 0.26.0. Everruns servers running more than one replica now claim each due session schedule atomically, so a scheduled message is sent once.
- [platform_management capability retired in favor of platform](https://everruns.com/docs/changelog/2026-09-11-retire-platform-management-capability/): 2026-09-11 · Platform, Cloud, Framework · Breaking · In 0.25.0. Capabilities gain a deprecated and retired lifecycle, and platform_management is the first one retired: it stays registered but no longer contributes tools.
- [Every Everruns user starts in a pinned Platform Chat thread](https://everruns.com/docs/changelog/2026-09-11-platform-chat-thread-every-user/): 2026-09-11 · Platform, Cloud · Feature · In 0.25.0. New users now land in a precreated, pinned Platform Chat thread instead of an empty Chats list, and Platform Chat gains more built-in capabilities.
- [Image generation adds the Muse image model with a fallback](https://everruns.com/docs/changelog/2026-09-11-muse-image-model-fallback/): 2026-09-11 · Platform, Cloud · Feature · In 0.26.0. The image generation capability can now use Meta's Muse image model through Meta or OpenRouter, and falls back to it when only those credentials exist.
- [Connect an LLM provider with one pasted key, checked before saving](https://everruns.com/docs/changelog/2026-09-11-llm-provider-quick-connect/): 2026-09-11 · Platform, Cloud, CLI · Improvement · In 0.26.0. LLM provider setup now offers a one-click grid of common providers, verifies the API key with the provider, and enables models and a default for you.
- [Everruns servers publish an MCP server card and auth.md](https://everruns.com/docs/changelog/2026-09-11-agent-discovery-documents/): 2026-09-11 · Platform, Cloud · Feature · In 0.26.0. Every Everruns deployment now serves an MCP Server Card and an auth.md document, derived from live config, so agents can discover how to connect.
- [AGENTS.md files resolve hierarchically as conversation context](https://everruns.com/docs/changelog/2026-09-10-agents-md-hierarchical-context/): 2026-09-10 · Platform, Cloud, Framework · Improvement · In 0.25.0. Agent instructions now load AGENTS.md from the workspace root down to the working directory and place them below harness safety instructions.
- [GPT-6 Astra, Claude Fable 5.1 and Muse Spark 1.3 model profiles](https://everruns.com/docs/changelog/2026-09-04-gpt-6-astra-fable-muse-spark/): 2026-09-04 · Platform, Cloud, Framework · Feature · In 0.24.0. Everruns adds model profiles for OpenAI GPT-6 Astra, Claude Fable 5.1 and Muse Spark 1.3, seeds them for new organizations and covers them in live tests.
- [MCP URL elicitation pauses for consent in the Everruns UI](https://everruns.com/docs/changelog/2026-09-03-mcp-url-elicitation-consent/): 2026-09-03 · Platform, Cloud · Feature · In 0.24.0. When an MCP tool asks a person to finish a step in the browser, the turn now pauses on a consent card and the secret never enters the model context.
- [Tool calls are validated against their schema before running](https://everruns.com/docs/changelog/2026-08-31-tool-argument-validation-strict/): 2026-08-31 · Platform, Cloud, Framework · Improvement · In 0.24.0. Everruns checks every tool call's arguments against the tool's JSON Schema and enables strict schemas for compatible OpenAI Chat and Responses tools.
- [Only organization admins can rename an organization](https://everruns.com/docs/changelog/2026-08-31-org-updates-require-admin/): 2026-08-31 · Platform, Cloud · Fix · In 0.24.0. Every PATCH /v1/orgs/{org} request now requires an organization admin, closing a path where members could rename non-default organizations.
- [Organization default model can fall back to the platform default](https://everruns.com/docs/changelog/2026-08-31-org-default-model-platform-fallback/): 2026-08-31 · Platform, Cloud · Improvement · In 0.24.0. An organization's default model is now an explicit override that admins can clear, falling back to a platform default shown in Models and Settings.
- [First message of a new chat no longer drops or shows out of order](https://everruns.com/docs/changelog/2026-08-27-first-chat-message-fix/): 2026-08-27 · Platform, Cloud · Fix · In 0.24.0. A new chat thread now replays from its first event when it subscribes, so the opening message is delivered and appears above later replies.
- [Everruns ID adds a sign-up page and branded account emails](https://everruns.com/docs/changelog/2026-08-27-everruns-id-sign-up-page/): 2026-08-27 · Cloud · Improvement. Creating an Everruns Cloud account now has its own sign-up page, and the verification and password-reset emails use the Everruns layout with a fallback link.
- [Messages API reports reasoning in order and the message phase](https://everruns.com/docs/changelog/2026-08-25-reasoning-ordered-message-phase/): 2026-08-25 · Platform, Cloud, Framework · Improvement · In 0.22.0. Session messages now carry ordered reasoning parts plus phase and phase_source, so clients can tell intermediate commentary from the final answer.
- [Global chat session endpoints removed from the Everruns API](https://everruns.com/docs/changelog/2026-08-25-chat-singleton-endpoints-removed/): 2026-08-25 · Platform, Cloud · Breaking · In 0.22.0. POST /v1/sessions/chat and POST /v1/sessions/chat/voice are gone; create a chat thread as an ordinary session with POST /v1/sessions instead.
- [Agents read several workspace files in one read_many_files call](https://everruns.com/docs/changelog/2026-08-24-read-many-files-tool/): 2026-08-24 · Platform, Cloud, Framework · Feature · In 0.22.0. The File System capability adds read_many_files, which reads 2 to 10 independent files in one ordered call and halves model round trips for discovery.
- [Session pages show a generated run summary and tab counts](https://everruns.com/docs/changelog/2026-08-22-session-run-summary-tab-counts/): 2026-08-22 · Platform, Cloud · Improvement · In 0.21.0. A finished run gets a one-sentence summary in the session header, and the Work, Events and Workspace tabs show how many items each holds.
- [Archive finished chats and show them again on demand](https://everruns.com/docs/changelog/2026-08-22-archive-chats/): 2026-08-22 · Platform, Cloud · Feature · In 0.21.0. Archiving hides a chat thread from the Chats list without deleting it, and a Show archived filter and the archive API bring it back when needed.
- [Provider connections carry custom headers and cache diagnostics](https://everruns.com/docs/changelog/2026-08-21-provider-connection-custom-headers/): 2026-08-21 · Platform, Cloud · Feature · In 0.21.0. An organization can set extra HTTP headers and a prompt cache diagnostics opt-in on a provider connection, applied to every model call made through it.
- [Start a chat with any harness, not only an agent](https://everruns.com/docs/changelog/2026-08-21-chat-with-harnesses/): 2026-08-21 · Platform, Cloud · Feature · In 0.21.0. The New chat picker lists active harnesses, including Generic, Base and custom ones, next to agents, so a chat no longer needs a placeholder agent.
- [Knowledge index embedding spend is metered and ledgered](https://everruns.com/docs/changelog/2026-08-14-embedding-spend-metered/): 2026-08-14 · Platform, Cloud · Improvement · In 0.18.0. Query embeddings made by search_index now count toward session cost and budgets, and index-sync embeddings are recorded against the organization.
- [Session recordings split into Transcript, Timeline and Events](https://everruns.com/docs/changelog/2026-08-10-session-transcript-timeline-events/): 2026-08-10 · Platform, Cloud · Improvement · In 0.18.0. A session page now offers three full-width views of a run: a read-only conversation replay, a curated execution history and the raw event ledger.
- [Chats is on for every organization, with pinned chats](https://everruns.com/docs/changelog/2026-08-10-chats-for-every-organization/): 2026-08-10 · Platform, Cloud · Improvement · In 0.18.0. Chats no longer depends on a feature flag, so every organization gets the chat list, search and new-chat flow, and threads can be pinned to the top.