> ## Documentation Index > Fetch the complete documentation index at: https://docs.daoco.org/llms.txt > Use this file to discover all available pages before exploring further. # Changelog > Product release notes for daoco. ## 1.1.8 Released July 22, 2026. ### Highlights * **Reddit research and engagement preview**: Workspace owners and admins can enable **Reddit** under **Settings → Preview features** to discover subreddits with fit scores, manage target communities, find threads worth joining, monitor brand mentions, and repurpose top posts from brand chat. daoco prepares a pre-filled post or reply for you to publish from your own Reddit account, then **Mark as posted** starts reply tracking in the Social Inbox. * **New personal and business brand onboarding**: Choose the type of brand you are building, then start with a public name, website, and social profiles or use 14 guided questions. daoco prepares an editable research-backed profile, fills gaps with targeted questions, and keeps website analysis running if you reload or visit another dashboard page. * **A better social research surface with broader source coverage**: Ask daoco to search specific supported platforms and see live progress while it checks connected APIs and the new public social-data broker. Completed searches show result counts and source categories for each platform, including clear partial-coverage and no-result states. ### Quality improvements * Social searches use connected native sources first, keep public-source searches bounded, and report unavailable platforms without failing successful results from other platforms. * Repeated research steps can reuse a matching completed search instead of triggering duplicate provider work. * Superseded website-analysis attempts cannot overwrite a newer onboarding run. * Reddit publishing stays in your hands, and broker-backed research uses public content without passing customer social-account credentials to the broker. ## 1.1.7 Released July 21, 2026. ### Highlights * **Faster, clearer brand chat**: Active turns now show **Queued**, **Understanding the request**, **Planning the approach**, and **Drafting** progress instead of leaving you with an unexplained loading state. * **Follow-ups that can steer work in progress**: Plain-text follow-ups appear immediately in the transcript and can guide the active run at its next safe step. If a run cannot be steered, daoco keeps the message queued. You can cancel a queued message or interrupt the active run when needed. * **Browser-backed work across more social platforms**: The Browser automation preview now supports authenticated UI tasks for LinkedIn, Instagram, Facebook, X, Threads, TikTok, YouTube, and Bluesky. daoco prefers native APIs, restricts browser work to approved platform origins, and retains step evidence and screenshots for review. * **More useful specialist results**: Delegated run cards preserve browser steps, links, screenshots, tables, partial results, and token usage without repeating the final narrative. Wide tables scroll horizontally on smaller screens. * **More dependable drafts and handoffs**: Requests for exactly one social draft no longer fan out into duplicate drafts, completed scheduling writes stay authoritative across specialist handoffs, and finished runs recover a visible result instead of ending silently. * **Cleaner sources and memory activity**: Article previews appear only when daoco extracts substantive source content, broken hero images stay hidden, and the timeline shows when background memory or workflow guidance was saved. * **Full-range campaign calendars**: Campaign workspaces render the entire campaign range alongside the selected week, with horizontal scrolling that keeps phases and dependency lines aligned. * **More connected creative and outreach tools**: Smartlead adds campaign, lead, deliverability, and sender-health workflows. Higgsfield and FLORA add provider-backed image, video, character, Technique, and creative iteration workflows through governed integrations. ### Quality improvements * Internal deliberation and provider lookup narration stay out of the customer-facing transcript while useful progress remains visible. * Superseded runs stop cleanly and cannot write stale customer-visible timeline updates over a newer run. * Browser enrollment rejects stale completions, keeps secure connection details out of the client and database, and records blocked target attempts in the audit trail. * Social publishing refreshes expired provider access more reliably, including recovery when scheduled or immediate delivery first encounters an expired connection. * Runtime completion, failover, and connection-evidence handling are more reliable for research, integration, campaign, and social workflows. * Large runtime prompts were tightened without removing workflow rules, reducing repeated context before specialist work begins. ## 1.1.6 Released July 2026. ### Highlights * **Pause, resume, stop, and retry in brand chat**: Active chat runs now have clearer controls. Pause asks daoco to stop at the next safe point, Resume continues a user-paused run, Stop ends the active run, and Retry starts a fresh attempt from the latest user request. * **Queued follow-ups while work is running**: You can send follow-up instructions during a long-running chat task. daoco queues them, shows them below the composer, and lets you edit or remove them before they run. * **More accurate billing for interrupted work**: Stopped, timed-out, and failed runs now settle the basic tokens already consumed instead of leaving holds open or refunding work that had already run. * **Cleaner chat timelines**: Specialist progress, delegated run cards, memory/tool output, article previews, and stopped-run states are easier to scan, with less low-value reasoning text in the transcript. * **More reliable published-post analytics**: Post and campaign analytics refreshes handle stale or orphaned data better, include clearer refresh status, and recover more reliably when provider metrics arrive later. * **Stronger scheduling and campaign views**: Campaign calendars, published-post placement, social queue grouping, and post publish side effects are more consistent across post, campaign, and chat workflows. * **Better brand refresh and onboarding**: Website refreshes allow slower crawls, logo discovery is stronger, empty source previews are suppressed, and no-op brand refreshes no longer show an unnecessary review panel. ### Follow-up updates * **Automations, the Social Platform Import Job, and the production integration categories are now available to every workspace**: these features graduated out of preview, so their opt-in toggles are gone from **Settings → Preview features**. * **Refreshed brand dashboard**: command center blocks were rebuilt with honest metrics (Published, Interactions, Seen), account snapshots with follower changes, real empty states, and direct links into posts, campaigns, calendars, and analytics. * **Clearer per-post analytics**: metrics a platform cannot report show as "Not reported" instead of zero, refresh errors read in plain language, and account metrics refreshes repair expiring platform tokens automatically. * **Team invites now require the Growth plan**, matching the workspace team model. * **Smarter multi-step chat routing**: daoco follows explicit instructions like "skip research", "only create the image", or a requested step order, and can apply saved workflow preferences when planning a turn. * Fixed mobile multi-image chat attachments and updated the underlying models across chat, asset, and analysis workflows. ### Quality improvements * Queued follow-ups keep the active run from being overwritten by a second run. * Paused delegated specialist runs now show clearer paused/resumed state. * Long specialist results are compacted into model-facing summaries with durable IDs and artifact references, reducing failures from oversized tool output. * Local and staging deployment automation is more predictable, including idempotent staging branch/tag handling and safer Convex environment sync. * Browser automation evidence is scoped and bounded more carefully when supported integration workflows collect page evidence. ## 1.1.5 Released June 2026. ### Highlights * **Social Platform Import Job preview**: Workspace owners and admins can opt in to a daily preview import that captures recent native posts from connected social accounts into **Manage Posts**. daoco groups likely cross-platform duplicates and preserves available metrics without publishing anything back to the platforms. * **Billed tokens in chat**: The composer context circle now shows how many basic tokens the thread has billed so far, alongside context pressure. Hover the circle for input, output, and total token details. * **More reliable LinkedIn publishing**: daoco refreshes LinkedIn OAuth tokens before scheduled and immediate publishes so connected profiles and Pages stay publish-ready longer. * **Cleaner scheduled publish state**: Scheduled social posts update auth and delivery state more reliably when publishing finishes, fails, or needs reconnect. * **Smoother asset uploads and exports**: Upload and export flows use more dependable storage handoffs for large files and downloads from chat and the Assets surface. ### Hotfix updates * **More durable agent runs**: Large tool results are now stored out-of-band before chat messages are saved, reducing failures when research, integration, or publishing tools return oversized payloads. * **Fixed Google Drive folder browsing**: Drive folder listings now resolve nested folder contents more reliably when using the Google Workspace connection. * **Stronger runtime status tracking**: Agent runs now record clearer timeline receipts, permission decisions, connection health, and no-artifact boundaries so chat and specialist work recover more predictably. * **Better production diagnostics**: Sentry events now include richer runtime, organization, thread, and request context to make production issues easier to trace and resolve. ### Quality improvements * Clearer context pressure warnings before automatic compaction. * More reliable export payload handoffs for asset downloads. * Improved R2 upload compatibility and credential recovery for media storage. * Stronger LinkedIn publisher error handling and scheduled delivery cleanup. * Updated token usage guidance and Social Platform Import Job preview docs in the knowledge base. ## 1.1.4 Released June 2026. ### Highlights * **Approval preferences**: Brand owners and admins can set per-brand defaults for scheduled social posts and integration writes, with **Always ask**, **Learn after 3**, and **Auto when connected** modes. Both policies default to **Auto when connected**. * **Automation prompt self-improvement**: Automations can optionally refine their saved prompt after successful runs. The legacy workflow memory library UI was removed in favor of the saved prompt plus an **Improve prompt** toggle. * **LinkedIn browser auth**: LinkedIn UI tasks can use a secure browser session alongside OAuth. Complete browser auth from chat, manage the session from Connections, and clear it without disconnecting OAuth. * **Browser automation preview**: Enabled workspaces can opt into browser automation for supported integration workflows, with clearer preview-feature gating in workspace settings. * **Smoother workspace invites**: WorkOS invitation links, invite acceptance, workspace selection, and team rescinds are more reliable for new and returning teammates. * **Clearer billing and top-ups**: Token top-ups now roll into billing totals, purchase history is easier to review, and inactive plans show remaining token balances. * **Dark mode in Account settings**: Switch between light and dark themes from your account preferences. * **Faster chat follow-ups**: Follow-up turns start faster, delegated specialist work stays in its own thread, and optimistic messages are more reliable after you send again. * **Knowledge base refresh**: Product docs now describe approval preferences, automation prompt self-improvement, LinkedIn browser sessions, specialist runs, workflow panels, and integration routing aligned with the current chat experience. ### Quality improvements * More reliable orchestrator intent classification under timeout and load. * Safer asset generation auth boundaries and memory extraction indexing in chat. * Attachment source menus close predictably after you pick a source. * Stronger approval workflow safeguards and orchestrator pause handling. * LinkedIn browser enrollment and session observability improvements. * Browserbase-backed social scraping replaces the previous Apify path. * Removed the legacy webpage builder product surface to keep the dashboard focused on chat, publishing, assets, and campaigns. ## 1.1.3 Released June 2026. ### Highlights * **New Growth plan**: Added a self-serve team plan for growing teams, with more monthly tokens, support for up to 10 brands and 5 seats, and discounted credit packs. * **Many more integrations**: Expanded integration support across commerce, website, mail, calendar, files, docs, work, knowledge, CRM, sales, and clearly marked Dev Testing providers. * **Clearer pricing and token guidance**: The knowledge base now includes billing and token usage pages so teams can understand plan limits, add-ons, top-ups, and usage. * **Sharper first-run experience**: Brand setup, the getting started guide, public site copy, and pricing flows have been tightened so new teams can understand what daoco does and start faster. * **Cleaner planning views**: Posts that go to multiple places are easier to scan, schedule, review, and act on from the planning workspace. * **Better mobile scheduling**: Scheduled posts are easier to read and open from smaller screens, with smoother schedule editing and publish progress states. * **A more visual asset library**: Browse images and media in a cleaner gallery with easier previews, selection, and detail views. * **Smoother chat experience**: Chat is easier to start, easier to read, and better at showing useful post, research, and workflow results without extra noise. * **Clearer publishing steps**: Drafts, schedules, media variants, and final approvals are easier to check before posting. * **Updated platform focus**: Platform support and preview-feature guidance now better reflect the currently available publishing, integrations, Pinterest, and social workflows. ### Quality improvements * Fewer confusing duplicate or stale messages in chat. * Clearer messages when something is not available yet or requires a connected account. * More consistent handling for drafts, attachments, media variants, schedules, and publish failures. * More reliable post compatibility checks before scheduling or publishing. * Better handling for integration reconnects, provider availability, and workspace-gated preview categories. * Cleaner analytics and observability paths across the app and backend services. * Streamlined preview-feature handling after removing unfinished automation surfaces from the active product. * Updated internal agent and runtime guidance so future work follows the current Convex-native stack. ## 1.1.2 Released June 2026. ### Highlights * **Refresh your brand from chat**: Start a brand refresh, review the suggested changes, and apply them only when they look right. * **Cleaner planning views**: Posts that go to multiple places are easier to scan, schedule, and review. * **Better mobile calendar**: Scheduled posts are easier to read and open from smaller screens. * **A more visual asset library**: Browse images and media in a cleaner gallery with easier previews and selection. * **Smoother chat experience**: Chat is easier to start, easier to read, and better at showing when work finishes in another thread. * **Clearer publishing steps**: Drafts, schedules, media, and final approvals are easier to check before posting. ### Quality improvements * Fewer confusing duplicate messages in chat. * Clearer messages when something is not available yet. * More consistent handling for drafts, attachments, media, and schedules. * Easier-to-understand brand update previews. ## 1.1.1 Released May 2026. ### Highlights * **Automations preview**: Enabled workspaces can schedule recurring prompts and review automated runs from daoco. * **Integrations preview**: Enabled workspaces can connect external tools and ask daoco to help with approved provider tasks from chat. * **Smarter chat routing**: daoco does a better job sending your request to the right helper for writing, research, publishing, SEO, campaigns, or visuals. * **Stronger publishing checks**: Drafts and posts get clearer review before publishing. * **Citations in research**: Research and chart answers can show where key information came from. * **Improved onboarding**: The Getting Started card became the main desktop guide, and website-based brand setup became more reliable. * **Reorganized docs**: Core workflow, billing, usage, and platform support guides became easier to navigate. ### Quality improvements * Clearer visibility for features that are not enabled yet. * Smoother imports, chat history, and specialist results. * Clearer supported-platform information.