Troubleshooting

Start by identifying the boundary that failed. Most d5s problems fall into one of five layers: access to the resource, acceptance of a message, schedule eligibility, capability authorization, or execution after a run started.

Slack is connected but my first reply is missing

Workspace connection alone does not add d5s to your channel or link every teammate’s account. In Slack, open the channel details, choose Agents & apps, and add d5s. Mention @d5s with your request. If a private Connect d5s prompt appears, sign in and finish linking before returning to Slack. Use the private Choose agent dropdown if the thread has no selection; only agents you can access appear. Ask a workspace admin if your agent is unavailable. The illustrated Slack guide remains available under Manage agent → Channels → Slack, with Open Slack below it.

An agent says its work is incomplete

If an agent sends a short message saying the work is incomplete, open its conversation in d5s. Review the completed actions and remaining work before asking it to continue. A limit can stop work before the full request finishes. When a run used all of its steps, the reply shows Stopped after N steps, the limit for one run. Select Continue, or reply continue, to pick up from there.

A message that is queued for approval has not been delivered. Check the outgoing message or email card before repeating a send. If delivery is unknown, check the destination first. A timeout or connection failure can also stop a run without a final channel reply.

First-login setup did not finish

If workspace setup or saving your welcome preferences fails, use the retry action on the welcome screen. Retrying reuses an already created personal workspace and does not rename it. If a region is unavailable for new workspaces, choose an available region. An invitation to an existing workspace does not require you to create a personal workspace.

Agent setup stopped at a connection step

Selecting connectors records which services the role uses; it does not connect an account. Choose an account, connect one as a workspace administrator, or select Set up later. A deferred connector is excluded from the agent’s access until you configure it later.

After Slack or mailbox authorization, return to the same setup draft. The corresponding channel settings reopen with the result or error; cancellation preserves the draft so you can retry. An inbox marked Enabled, a connected Slack or Telegram identity, or a Discord identity (beta) does not mean the draft can run. Review the draft and select Activate agent before testing messages.

If activation reports a conflicting connector authentication method, change the selected connections and retry. You do not need to create another draft or channel identity. Your setup history remains available after activation in Agent settings → Setup history.

An agent cannot find a Slack attachment

Mention the agent in the same thread as the file and ask it to inspect the attachment. The agent can discover files in earlier thread messages, but thread history and attachment counts are bounded. If the context says content was omitted, attach the needed file directly to your new request. Files that Slack no longer makes available, or that exceed the supported size limit, cannot be downloaded.

No-training work is blocked before or during a run

First check Settings → Preferences → Privacy. If Use no-training routes for my work is on, your interactive and owned background work requires the no-training route. Also check the organization's Privacy page with an Owner or Admin: Require no-training routes for everyone applies even when your personal switch is off.

A Slack message is a third case. A Slack message to someone else's agent follows the stricter of your and the owner's no-training settings. Your turn can therefore require the no-training route even when your own switch is off. If no eligible model is available, the agent replies in the thread and clears its own status. Ask the agent's owner to check Preferences → Privacy.

If no-training work cannot start or reports no eligible provider, keep the privacy requirement in place and contact d5s with the selected model and visible error. Retrying cannot bypass the requirement. Turning off your personal setting returns to normal routing only when the organization does not require no-training.

Opening Settings from an older view

In the browser and d5s Desktop, opening Settings can automatically bring an older view up to date and continue to the page you selected. You do not need to repeat the click or choose Reload for a normal release. When you close Settings or go back, you return to the page you left with your unsent draft, attachments, selected chat options and position restored.

Update d5s Desktop to use the same behavior when opening Settings from its app menu.

An open Settings form stays stable while you edit it. If another editor is open or a save is still pending, finish its existing Save or Cancel flow before the update can continue. d5s does not submit or repeat your work as part of an update.

A page could not load

If d5s shows Reload when you’re ready to continue, a page could not load. Your current view stays open so you can save any unfinished work before choosing Reload.

An unsent chat reply and its attached files stay on this device after a reload or a closed tab. They come back when you open the chat again, for up to 7 days. Sending the reply or clearing the box removes the saved copy. Signing out removes all of them.

In the browser and d5s Desktop, Reload continues to the page you tried to open. If it still cannot load, check your connection and try again. Report the problem if it continues.

I cannot sign in

Open d5s and continue directly on the hosted authentication screen. It offers the sign-in methods enabled for d5s, including Google, GitHub, and email. Complete any email-verification, multi-factor authentication, or organization-selection step there before returning to d5s.

Public desktop releases currently open EU automatically, including when an older release saved US as its preferred region.

In d5s Desktop, sign-in uses your default system browser so you can use the accounts and profiles already signed in there. Desktop first shows Sign in to d5s and does not open another application until you choose Continue in browser. Keep d5s Desktop running, finish the browser flow, and allow the browser to open d5s when prompted. The browser then shows Browser sign-in complete and normally returns to the app automatically; choose Open d5s if it does not. You can close that browser tab after the app opens. While it waits, Desktop shows Finish signing in in your browser; choose Open browser again there if the sign-in tab was closed or lost. Desktop reopens a live attempt or starts a fresh one automatically if the old attempt has expired. If the app still does not return, quit and reopen d5s Desktop before starting a new sign-in; an old or already-used return link cannot resume a new attempt. After you sign in again, Desktop keeps only the current tab and closes the others.

If d5s returns to the login page, follow the specific message shown there. Verification, MFA, additional identity checks, and organization selection must be completed in the hosted authentication flow, so start sign-in again and finish the named step. If the message says your d5s account is inactive, contact your workspace administrator instead of retrying.

On a shared device, sign out of the identity provider separately before signing in to d5s as another account. The d5s signed-out screen confirms only that the d5s session ended.

If a challenge repeatedly returns to the login page, send support the approximate time of the attempt. d5s support will never ask you to paste a verification code, pending authentication token, or OAuth token into an issue.

An approved video is still processing or has an unknown outcome

If the agent reported that the video is processing in the background, it can finish after the agent's reply. Check its approval card and return to the conversation later. Do not approve a second copy while the first is processing.

If the outcome is unknown, use Report a bug and include the conversation and approval time. The video may still exist, and another generation can incur another charge. If the card reports a confirmed failure and offers Retry, that action starts a new generation.

A run will not start

Check the visible session state first.

• If a turn is queued, the session already owns active work. Wait, or cancel the queued turn.
• If the session is active, stop it only when you intend to interrupt the current tool work.
• If an agent is paused or its daily budget is spent, manual and autonomous work can be blocked until resumed or until the next local day.
• If an automation already has a run in flight, another occurrence is skipped instead of overlapping it.
• If the response reports a credit block, the box you sent from replaces its send button with a message that names the cause: an empty balance, a full pace window without purchased credits, an organization or workspace spend limit, the member's spend limit, or no active plan. It gives the time work resumes when one is known and presents available billing actions to an organization Owner or Admin. A Using purchased credits notice means metered work can continue after a pace limit is reached; purchased spending does not fill that subscription window. Read the refusal message as the authority on whether a turn can start. See Plans and credits.
• If the request is rejected before a session appears, verify workspace access, project access, model policy, and uploaded file ids.
• If a newly listed model is missing from the composer or cannot be selected, ask an organization Owner or Admin to review its availability under Organization settings → Models and enable it when the account can use it. New model families are not enabled automatically. New versions of an enabled family become available automatically when they meet the account's plan, provider and privacy requirements. If no eligible version remains, choose another available family or ask an administrator to review the settings. If the model cannot be enabled, contact d5s with the organization name and visible error.
• If d5s shows Choose a plan to start or Your subscription ended instead of the workspace, the account has no active subscription. Choose a plan, or accept a team invitation from your email. A banner with Renew subscription means the subscription ended and access stops on the date shown. See Plans and credits.
• If every model is unavailable in a new organization, ask an organization Owner or Admin to check that the Light level is available. Free organizations run the Light level only; Medium, High and hand-picked models require a paid plan. Contact d5s with the organization name if an expected model is unavailable.

If a workspace is missing from the switcher and an Other regions section is shown, use Open region to check the region named in your invitation or onboarding instructions. A region that is not open yet is not listed. If a direct link instead says Workspace not found here, use Open in region or Other regions. When no other region is available, use Choose a region. The requested page path is preserved, but query parameters and fragments are removed and you may need to sign in to the destination.

d5s deliberately does not claim that a neutral region link contains your organizations. The destination asks you to choose an account on its hosted sign-in screen; select the account named in the invitation. After the regional app opens, look for the organization and workspace there. If either remains missing, ask an organization Owner or Admin to assign you or one of your groups from the organization's Workspaces page. An organization Owner or Admin, or an Owner or Admin of the workspace itself, can instead open Settings from the intended workspace and invite you by email from People → Invite people; the workspace's own Members section grants access to people who are already in the organization, and opening that section needs an organization role even for a workspace Owner or Admin. Organization membership and a workspace grant are separate; joining the organization as a plain member does not grant access to all of its workspaces.

If Usage → Workspace stats shows a Historical workspace label, the usage entry still refers to a workspace ID whose current name is unavailable. This preserves the organization's totals and does not indicate that the workspace can still be opened. The table covers the complete selected period; its chart says when it is limited to the most recent 366 days.

Understand what an agent saw during a billed run

An organization Owner or Admin can open Usage → Activity, select an available row, and choose Inspect run. Review the billed outcome, recorded activity, model usage, and any failure shown. Use Open conversation to jump to the related turn when it still exists.

A run record may not contain the exact input sent to the model provider. Missing evidence is not a reconstruction of that input. Billing access does not grant access to conversation content, and a record is unavailable when the conversation has been deleted or you no longer have permission to view it.

A run may pause without new visible output before it finishes. Before retrying a failed run, check its recorded activity for completed external actions so you do not repeat them unintentionally.

An eligible chat or agent run can continue automatically after a service restart or unexpected interruption. You do not need to resend the request while recovery continues. Recovery does not resume work you stopped or grant access you no longer have. If the agent cannot verify an interrupted action’s outcome, it asks before repeating it. If you see The agent stopped before it finished., check recorded activity before retrying. Completed actions are not undone by the failure. If the error instead says why model recovery stopped, select Continue to resume from the saved work. Completed actions do not run again.

A result is shortened or part of a review is missing

Ask the agent to continue reading the available full result before giving a complete assessment. Long text can be read in smaller parts, even when it contains one very long line. If the full result is unavailable, ask for a narrower read from the original source. A preview or an unavailable result cannot confirm that nothing needs action. Do not repeat an action that sends or changes data just to recover its output.

A run stopped because commands did not run

If the agent stops after repeated command failures, the run shows that it did not finish. Your workspace files remain available. Send your message again to continue. If this keeps happening, use Report a bug.

A sandbox command says linked credentials are unavailable

A linked secret can expire, be revoked, or lose access to its workspace. This can block a command even when the script does not use that secret. An administrator can open Secrets & variables in the agent or project settings. Check the named link, restore access to its source, or remove the link. Removing a link keeps the workspace secret. Then retry the command.

If your script cannot find a credential, check its environment name and link. Sharing a secret with the workspace alone does not supply it to scripts. Do not paste its value into chat. See sandbox credentials.

A Python project needs another version or a timezone is missing

Ask the agent to read the project's Python version requirement before installing dependencies. It should create a separate project environment and run the project with that environment's Python. For example, a requirement of >=3.12,<3.13 needs Python 3.12. Ask it to leave the default Python unchanged.

For a timezone error, ask the agent to check the named timezone with the same Python that runs the project. A successful check in another environment does not verify the project. If an installation needs network access, use the normal approval process. Keep the error and conversation link if setup still fails.

A saved artifact will not open

Open the artifact from its card in the conversation. The card keeps the presented version, even after the coding environment stops. Files opens the current workspace file while it is available. When the coding environment stops, d5s also saves each new or changed file in the artifacts folder. If those files have expired, Open and Download use the latest saved version. Hidden files, dependency folders such as node_modules, key files, and files larger than 25 MB are not saved. If a saved card still fails, use Report a bug and include the conversation link.

If the preview shows Preview did not load, the file is safe. Select Retry. After three retries, wait and open the file again later. Download stays available while the preview fails.

Continue coding after the environment restarts

Starting your coding environment… means the agent is preparing its coding environment. Wait for the activity to continue before treating the application as ready. You can stop the run while it is starting.

If an older retained environment cannot be migrated safely, startup stays unavailable. Contact support to recover its saved work. Retrying does not reset it.

Ask the agent to inspect the saved project and continue from its files. Open the saved project under App preview if it appears there. Otherwise, open its conversation tile or find it under Browser. A running app can stay viewable after the agent finishes. Select Start when its environment or server is stopped. With permission to run work, open ⋯ → View startup logs to inspect startup output or ⋯ → Restart app to replace a server started through Preview. If Start is unavailable, ask the agent to start the app. A saved project keeps any private website browser under its existing control. For missing dependencies or an environment busy with other work, ask the agent to inspect and start the app.

A retained coding environment keeps its project files, installed dependencies, and workspace settings while that environment remains available. Saved artifacts remain available after that environment expires. After expired working state has been retired or migrated, a later sandbox task can start in a clean environment. Restart can remain unavailable while cleanup completes. Open saved results from Files or their artifact cards. Saving an artifact does not save every other file in the coding environment. Other files created only in the coding environment need a separate saved copy. An environment rebuilt from saved files does not retain dependency folders such as node_modules and .venv or regenerable caches. Ask the agent to inspect the environment and reinstall missing dependencies from saved manifests before continuing.

Newly saved coding work preserves executable permissions and supported file links. An older saved copy may have lost them. If a script reports a permission error or a tool is missing, ask the agent to repair or reinstall it. Restarting alone does not fix this.

If your shared computer is still running, ask the agent to inspect its current work before restarting it. A lost viewer connection alone does not mean its running commands have stopped. Reopen the computer to check its state.

While the computer stops, agent commands wait for the stop to finish, up to about three minutes, and then start the computer again. If the Computer panel says Computer is stopping, a stop has taken more than two minutes. A workspace owner or admin can select Restart computer. This stops running commands and closes open desktops for everyone in the workspace; files in the workspace stay. The computer starts again when someone opens it or an agent needs it. A file that an agent saved just before the stop stays on the computer, but it may not appear as a saved version; ask the agent to save it again. If the panel says Computer could not stop, contact support.

If your shared computer restarts automatically, wait for it to become available before continuing. Its saved files remain, but running commands and open application connections do not resume. Inspect the last result before repeating an action. A computer you explicitly stop stays stopped. If restart cannot complete, contact support; do not reset or delete its saved work.

Commands fail after a file upload

Uploaded files remain read-only inputs. Save changes as a new working file or artifact. Shared workspace computers support individual uploads and downloaded chat-channel files up to 100 MiB. Email attachment limits still apply separately. If the result says the request exceeded a size limit and the command did not start, use a smaller input before trying again. If the outcome is unknown, do not repeat the action until support confirms its result. If an existing upload prevents commands or screenshots from running, contact support with the conversation link. Keep the original file; deleting uploads is not a recovery step.

Home looks empty in the computer's file manager

Select Workspace under Places to return to the chat's working files. Files opens that folder by default. Home contains private application settings, which are often hidden. Selecting Home does not mean your workspace files are missing.

Computer shows Desktop unavailable

The viewer could not connect to the desktop. This does not confirm whether the last task succeeded or failed. Select Retry when it is available, then inspect the current state before repeating an action that changes data. A terminal command can finish even when the viewer is unavailable.

If retry does not restore the desktop, report the problem from the conversation and include the time and visible error. Do not repeat purchases, form submissions, or other changes solely because the viewer disconnected.

Computer desktop opened without its earlier tabs

A project desktop closes after about 10 minutes with no viewer, command or desktop action. The next time you or an agent opens the computer, the desktop starts again. Saved files remain, and the browser normally keeps its sign-ins. Open tabs, windows, terminal sessions and unsaved app input do not. Reopen the page you need, or ask the agent to continue from the current screen.

Terminal is not open when the desktop starts. Select it in the dock to open it.

Computer storage is full or unavailable

If workspace settings show Storage, a workspace owner or admin can open it and select Refresh. Storage is full means you must free space before saving new files. Check the files you plan to remove, especially when other work in the project uses them.

Usage is unavailable while the computer is stopped or unreachable; that is not a report of zero usage. Refreshing the page does not start the computer or increase storage.

An app preview is unavailable or temporarily busy

Open the source conversation. Choose the saved project under App preview if it appears there. Otherwise, open its conversation tile or find it under Browser. Ask the agent to show the app again if it is still unavailable. If a preview asks you to sign in, sign in to d5s again and reopen it from the conversation. This also applies to previews in separate tabs.

If the preview is temporarily busy, close previews you no longer need and retry. If you can view a project but cannot use ⋯ → View startup logs, ask its owner to check your permission to run work. Viewing an app does not grant that permission. For an app that is stopped, use Start when available, or ask the agent to start it.

The coding environment reached a storage limit

If the coding environment is running, ask the agent to inspect its storage usage and remove generated dependencies, caches, or temporary files that it created. Keep your uploaded files, source work, and final outputs. Commands or file writes that would continue growing an oversized environment are refused; an active command can also be stopped after it crosses a limit. Saved work is preserved, and cleanup does not happen automatically.

After the agent removes the selected generated files, it can retry the task. Repeating the same installation can reach the same limit again. Restarting does not increase the storage budget. If the task still needs more space or files, ask the agent to explain that requirement before continuing.

A storage limit does not prevent restarting to inspect saved work and remove selected temporary files. Older saved work can exceed the current budget. Inspect and clean up that work before continuing. If restarting fails, report the error through Report a bug.

If a dependency install or build is killed or reports that it ran out of memory, ask the agent to reduce parallel work or process smaller batches before retrying. Repeating the same memory-heavy command can hit the limit again. If the task still cannot fit, report the command and error through Report a bug.

An image edit cannot use the source picture

Make sure the picture is still attached or available in the conversation, and name the exact image you want to change. Editing accepts PNG and JPEG sources up to 3 MiB, 4096 pixels per side, and 4 megapixels. For a larger picture, an unsupported format, or an animated image, ask the agent to prepare a separate supported copy or a single frame. If the file is missing or damaged, attach a readable copy before retrying. A failed source check does not start image generation.

An image is too large to inspect

For an otherwise supported PNG or JPEG that exceeds the pixel limits, the agent automatically creates a smaller temporary copy and reads it. The original stays unchanged. If the image still cannot be processed, ask for a smaller copy or a crop of the relevant area. Resizing does not add image-reading support to a model that lacks it.

An Office conversion or spreadsheet calculation fails

Ask the agent to convert the original Word document or PowerPoint deck to PDF and inspect the rendered pages before delivery. Keep the source file. Recreating its text can change the layout.

For an Excel model with empty or stale formula results, ask the agent to recalculate a copy and verify the saved results. An empty formula result is not zero. Macros, external data, add-ins, or unsupported formulas can require Excel itself.

If the agent reports that conversion or recalculation is unavailable, use Report a bug. Include the conversation link, file type, and visible error. Do not repeatedly request package installation or discard the original file.

A Word preview looks different

When a Word page preview is unavailable, d5s shows a simplified preview that can differ from the original formatting. Saved checkbox selections remain visible; wide tables can scroll horizontally. Download the original file to compare it in Word or edit form answers. For files with tracked changes or comments, use Markup to review them and Final to see the clean result.

A scanned PDF or image has no extractable text

A scanned page can contain a picture of text without an embedded text layer. Ask the agent to read the PDF with its PDF reader; it reads scanned pages without OCR. For a long scan, name the pages you need. If PDF reading is not available for the selected model, ask the agent to use OCR for English, Dutch, or Greek, and name all languages present. Review extracted names, numbers, and table structure against the source. For visual layout or diagrams, use a model that supports image reading; OCR does not add vision support to a model.

A message send lost its response

Open the same conversation and inspect its recent history before sending a new copy. A connection failure can happen after d5s accepted the message, and acceptance does not mean the agent has finished.

For an API integration that assigned a client request ID to a persistent agent or regular chat message, retry with the same user, workspace, session, ID, and exact original payload. Do not edit the payload or generate a fresh ID just because the first response was lost. Follow the recovered turn in history. If the server reports a conflict or rejects keyed sends for that session, inspect the request and use the developer retry guidance before trying again. This contract does not change the separate retry rules for schedules or external message delivery.

A website replaced the page I wanted to keep

Ask the agent to open the next website in a new tab and then list both tabs. Navigation changes the selected tab. If the agent reports success but you see one tab, ask it to inspect the current tab list before trying again.

If a browser operation is interrupted, inspect the current browser state before repeating it. The operation may already have changed the page. Use the browser's recovery action if the connection was closed.

An agent is waiting for browser access

For Waiting for website approval, select Show request in chat. The Allow a website? card names the exact website the task needs. A workspace admin reviews the reason and selects Allow and continue or Decline. Current browser access shows the existing allowed websites. Sign-in can need more than one website; each new hostname needs approval. Allowing a website does not share credentials with it.

If website access was saved but the task has not continued, select Continue task. An expired or changed request needs a fresh request from the agent. A failed save leaves the request available to retry. Private control and deliberate pauses still need you to return control.

An agent with approved browser access can start its saved browser under either conversation approval setting. Starting a saved browser can take several minutes. During Starting browser, the agent keeps the original request and continues when the browser is ready; opening the panel is optional. If access changed, review Browser settings, then ask the agent again. Private control and deliberate pauses still require you to return control.

For Private control, choose Return to agent when you are ready. For a paused browser, choose Resume browser. A deliberately stopped browser needs Open browser. If another conversation is using the saved browser, review its use in Browser settings. A startup failure shows its recovery action; changing control during startup requires reviewing the current controller before returning control.

This agent is archived means the browser cannot run until you restore that agent. For other connection failures, use Reconnect browser or Retry. If Browser access is off, open Browser settings to enable it. Existing Off choices and website restrictions do not change with the new defaults.

A workspace admin can use the approval card in the agent conversation or open Browser settings. If the site was approved but the browser did not start or return control to the agent, check that profile in Browser settings. If opening an app in the coding sandbox is denied, ask a workspace admin to enable Allow browser in Browser settings; an automatic saved browser does not need a separate profile choice; approving a public website does not grant the app port. If the profile is ready but the conversation has not continued, choose Ask agent to retry. A request that changed while you were approving it needs a fresh review in Browser settings.

An invoice PDF did not arrive

For a Moneybird invoice PDF, first check that the agent used the intended connected account and that the download action was allowed or approved. A supported download should import the PDF during that connector call, without a separate website-access request.

If the agent reports a file import error, the temporary link may have expired or the provider may have returned an error or a non-PDF response. Ask the agent to inspect the saved connector result before retrying the read-only download action. If Moneybird reports a retry time or rate limit, wait until then. Do not repeat a write action to recover a file. For other connectors, see using connector results.

A connector action is unavailable

Open the credential's Tool permissions with Manage permissions on its connector detail page. Find the requested action and check its choice: Allow, Ask first, Block, or the default.

Check the connector chain in order:

1. The workspace credential exists and is not archived or awaiting reauthorization.
2. The connection is available to the project or agent doing the work. For an agent, inspect Manage agent → Connectors.
3. The requested tool is Allow, an Ask first approval is still live, or the session is visibly in Allow all mode. Workspace-secret Custom API requests still need Approve once in every mode. Gmail and Outlook sends and calendar bookings always wait for an approval card, and they are refused in automations, dashboards, and sub-agents.
4. The requested action is within the connector's supported access.
5. The external account can access the requested object.
6. The provider account has sufficient credits, quota, and feature access for the requested operation.

When several connections use the same provider, confirm that the intended named connection remains attached and that its stable handle is the one the agent used. If an Attach card says the connection is no longer available, refresh the conversation and search again; the credential may have been archived, moved into reauthorization, or already attached since the suggestion was created.

If an Ask action is waiting, return to the conversation and answer the pinned connector approval card. Approving once does not approve later calls. You can instead switch that conversation to Allow all, which retries the pending action once. If the retry fails, the approval card remains available. If Allow all is not shown or an action still pauses, the feature may be unavailable for the workspace or an explicit Deny or another connector boundary may be blocking the call.

The provider changed its tools

If the agent reports that a tool definition changed, have it inspect the same connected account again and check the current operation arguments. Do not switch accounts to work around the error. A discovery error is different from a successful empty tool list. An administrator can refresh the credential's reviewed tool inventory; a failed refresh keeps the previous snapshot.

If an operation times out or its output cannot be saved, it may already have executed at the provider. Check the provider's state before repeating a write. A returned resource link is a reference, not proof that its target has been downloaded.

If a connected account remains unavailable, check its authorization and reconnect it if requested. If the error persists, contact support with the connection name and visible error.

For Jira, confirm the intended connected Atlassian account and site. Ask the agent to list accessible sites and choose one explicitly when several are available. Reconnect the exact account if it needs reauthorization. A missing project, issue, transition, or permission must be repaired in Jira; repeating a d5s approval cannot grant that access. Files larger than 50 MiB cannot be uploaded or read through this connector. If a result import fails, inspect the saved result before requesting another operation. After an uncertain comment, issue change, link change, or upload, check Jira before retrying because the change may already have succeeded. See Connect Jira Cloud.

For HubSpot, check the exact account handle and action the agent selected. Reconnect that account if it needs reauthorization; connecting another account does not repair the first one. A former HubSpot MCP connection is not converted to the new OAuth connection; disconnect it and reconnect HubSpot. Reattach the new connection to the relevant projects or agents. For a new HubSpot connection, inspect its available tools, then check the action policy and confirm that HubSpot consent included the matching CRM scope and product access. If a write's result is uncertain, inspect the records in HubSpot before retrying; the action may already have executed.

If a remote MCP connection fails before the provider's sign-in or consent page appears, retry once and then contact support with the provider name, approximate time, and visible error. The provider may be unable to read d5s's public client metadata or may not accept its client domain. Never include an OAuth code or token in the report.

A connector upload is too large

One connector request can send at most 100 MiB. A larger request fails with request_body_too_large. Waiting and retrying does not help, because the limit applies to each request, not to a time period. Ask the agent to compress the file, split it into smaller files, or use the provider's resumable upload when the connector supports it. MCP tool calls have the same 100 MiB limit. Smaller connector limits, such as 50 MiB for a Jira file or a downloaded connector file, still apply. A request_body_buffer_busy error means that other large requests are in progress. Nothing was sent, so retry after a short wait.

An agent does not use a workspace skill

An agent uses only the workspace skills on its own Skills list. Every agent gets each workspace skill automatically. If the skill is not on the list, someone removed it from the agent. Open the agent, select Manage agent → Skills → Add skill, and add the skill or its plugin again. See Choose an agent's skills. In a project chat, the project must also have the skill turned on: open the project's Context → Skills and turn the skill on.

A scheduled run did not happen

For an automation, verify enabled state, trigger kind, timezone, next due time, daily quota, and whether a run is already active. Missed recurring times run once when the automation can run again. A one-time occurrence runs when it can, then turns off.

For an agent scheduled wakeup, verify its enabled state, the agent pause state, timezone, next run, last status, and daily budget. A busy or temporarily failed occurrence runs again later. A one-time wakeup also remains due after a budget skip and retries when budget becomes available. If the wakeup's run fails for a temporary reason, such as a busy model, before the agent takes any action, d5s retries it automatically a few times within the next hour, and the agent knows the run is late. A disabled wakeup, or a new time you set, cancels the retry. A one-time wakeup whose run failed without a retry turns off but is not marked complete; set a new time to run it again.

For a scheduled mailbox check, No new messages means no agent run is needed. Check Last check and any connection error. Follow mailbox check recovery before replacing a check whose mailbox changed.

Remember that Wake now does not move an agent's scheduled wakeups and Run now does not reset an automation's recurring schedule.

An agent did not enter a scheduled call

Open the agent's Meetings → Auto-join tab. Confirm that the intended calendar is selected, the joining rule covers the event, the event is accepted when the rule requires that, the event has a Google Meet, Microsoft Teams, or Zoom link in its video call details, location, or description, and its individual switch is on. If the meeting came from another calendar, add the agent address shown next to Invite as a guest; the event must also appear on a selected calendar.

Then open the meeting under History. Waiting for host means the listener is still in the lobby. No one joined means the listener reached the lobby or the call but heard no speech before the call ended. If nobody admitted it, ask the organizer to admit the listener or include its address on the invitation before retrying. Declined by host means the request was refused. Call failed is a provider failure before the listener reached the lobby, or after it had captured speech, so verify that the meeting link still opens and that the meeting service is available. A transcript cannot begin until the listener enters the call.

Left empty room means the listener captured speech, then heard no one for 10 minutes. It cannot see who is in the call, so it treats a long silence as an empty room and leaves so the agent does not sit in an empty meeting. Muted participants count as silence: if you wait muted for more than 10 minutes, unmute briefly or rejoin the listener later. The transcript up to that point is kept. If the call was still going on, rejoin it from Join call and the agent returns as a new call. A whisper asked during a call is answered in the Meetings panel or in the meeting view; it is never a message in the agent's conversation, so do not look for it in the timeline.

Open the meeting view to read a call's notes: select the call in Live, or Open meeting on an ended call. If no one spoke and no one added a line, the call has no notes, and the notes pane reads No notes for this call. Add to notes, the whisper box, and Leave call each need their own permission; a person without it does not see that control. If a whisper answers Whisper is not available for this workspace's models, no light model is available to the workspace; ask an organization admin to review the allowed models, and use Ask agent meanwhile.

If Edit and Add to notes are missing while you have edit permission, someone else is editing the notes, and the notes bar names that person. Wait until they select Done; editing also ends after two minutes without input. If a save shows Could not save your last changes, your edit was ended, for example after a lost connection. Select Edit again and redo the last change. If the bar says the agent paused, the owner has no credits or the agent reached its daily budget; the agent continues after credits are added or on the next budget day. If it says the agent could not update the notes, it tries again on its next pass, and people can still edit the notes.

If the transcript shows the wrong language, or lines in languages that nobody speaks, set the language control in the meeting view header to the languages of the call. New lines use it within a few seconds; lines already written do not change.

An automation completed but its agent is still waiting

Check the execution status and each agent's handover status separately in Latest run or History. Pending handover means that agent has not received the result through an eligible conversation yet. Received confirms receipt, not human review or follow-up work. A successful run can still be waiting for an agent, and a failed run can have a received handover.

Select View run on the automation card, or open the automation and choose Open run or Open, to inspect the complete result directly. If the handover is unavailable, use Retry; do not treat a loading failure as confirmation. People with access can inspect and review the run without an attached agent.

Agent email did not send

Open the agent conversation and inspect the outgoing email card first. For a single recipient blocked by the mailbox's Sends to rules, the agent may offer an exact-address request in d5s or its supported originating channel. A d5s channel manager or a current member of the bound Slack or Teams channel can decide it. Telegram sends a d5s review link, and an email-origin request sends that link to workspace admins through platform mail. After allowing the address, ask the agent to retry; the blocked send does not resume automatically. Under Manage channels → Email, select the sending mailbox and check its Sends to rules. Confirm that the agent is not paused, its AI budget remains available, and the recipient is not suppressed. New email needs a selected or default mailbox; replies stay with the mailbox that received the original message. New automatic d5s inboxes require approval before sending. On a pending email card, a channel manager can choose Approve & send for that email or More options → Approve & always allow this recipient to exempt that exact address from later approvals on this mailbox. Existing policies may also require approval or restrict the agent to replies until explicitly changed.

An email activity state of Sent means its mail provider accepted the request; it does not prove recipient delivery. A timeout after submission is left Uncertain instead of being retried automatically, because a blind retry could send a duplicate. Failed and blocked entries include the reason recorded by d5s.

If an incoming email card shows Waiting for credits, add credits or wait for the usage limit to reset; the agent then starts the email automatically. If incoming mail does not start agent work, check that the receiving mailbox is active and its Receives from setting allows the sender. An email held only by the Receives from setting offers Allow sender & deliver this email on its card; a channel manager can allow that exact address and release the retained message once. Older blocked messages may offer Allow future email from sender and need the sender to resend. Under Advanced → Blocked senders, a matching block overrides all incoming permissions. When Accept replies from people this mailbox emails is on, an address this exact mailbox successfully sent to can also email the agent; drafts, failed or uncertain sends, and another mailbox's history do not qualify. This exception does not grant outgoing permission. When Review unknown senders first is enabled, successful-send history still does not bypass review; only entries shown under Trusted senders do.

For Gmail or Outlook, confirm the message reached the connected inbox and reconnect if mailbox authorization has expired. Allow time for the next mailbox check. Automatic responses and common bulk or mailing-list messages start ordinary agent work only when their From address explicitly matches the mailbox's saved Receives from rules; the successful-send reply exception is not enough. For a d5s address, inspect Needs review in Email activity. If the card says Sender could not be verified, read its DMARC explanation and ask the sender's email administrator to check SPF, DKIM, and DMARC. A channel manager can choose Deliver once, or open More options for a persistent sender or mailbox allowance after confirming the spoofing risk. Spam, virus, malformed-message, and blocked-sender checks remain blocked. An explicitly allowed message may use a different Reply-To, but replies remain addressed to its From contact. Read Agent email for setup and address-rule details.

If an incoming email attachment is not read, inspect its row on the email card. Unsupported or unsafe files are skipped with a reason. Supported files are fetched only when the agent needs them, so a reference can remain unread without indicating a failure. If a previously readable file now fails, confirm that the original message and attachment still exist in the same connected mailbox and reconnect that mailbox if its authorization has expired. For a d5s address, an unread attachment can become unavailable after the original message's 30-day retention period; supported files already downloaded remain under the agent's retained file controls.

Slack did not ask which agent should answer

Invite @d5s to the channel and select it from Slack's mention suggestions when sending your request. If a thread already has an agent, or the channel has a default, d5s uses that choice without asking again. Use Change agent in the thread to choose someone else. A thread choice takes priority over the channel default.

If d5s asks you to connect your account, use the private Connect d5s action with an account that can access the agent. Only the requester can use their agent picker. If the picker expires, mention @d5s again. If an agent is missing, confirm that you can open it in d5s and that it is available for use in this channel.

The optional shortcut is @d5s +poppy investigate, with no space between + and the agent name. It does not autocomplete. Unknown or ambiguous names are refused; use the picker instead of guessing a handle.

Shared Slack replies fail or still look like @d5s

The reply body shows the selected agent's name and character so the identity remains clear when Slack groups messages. The native Agent name · d5s sender also needs Slack's customized-message permission. For a connection installed before that permission was added, a workspace admin must open Manage agent → Channels → Slack and choose the Reconnect Slack link under Open Slack once before customized sends can resume. With several connected Slack workspaces, the sentence names each workspace as its own link.

A Slack coworker ignores replies in a thread it follows

A coworker installed before thread follow shipped still runs its older Slack app configuration, so Slack does not send un-mentioned thread replies to it. Such a coworker does not promise to follow: its first reply in the thread says it cannot follow the thread yet and links to Upgrade permissions in d5s. Follow that link, or open Manage agent → Channels → Slack, where a line under the identity names the pending upgrade. A workspace admin chooses the highlighted Upgrade permissions button, continues through Slack OAuth, and approves the added permissions; the existing app, channel connections, and history are retained. Threads that received the prompt are not followed retroactively; mention the coworker once more in each of them. Then confirm Thread follow is on for that channel, that the thread was quiet for less than 72 hours, and that nobody reacted to one of the agent's messages in the thread with :no_bell:. A fresh mention in the thread starts the follow again.

A Slack coworker message did not arrive

If the text message itself is missing, inspect the outgoing activity before asking the agent to send it again. Accepted by Slack means Slack accepted the API request; it does not prove teammate receipt or a read. Unknown means the request may have crossed Slack's acceptance boundary, so d5s avoids a blind retry that could create a duplicate. Check the target Slack channel or thread for the original message first.

An Unknown entry may show that another send attempt is pending after Slack refused the first one. Pausing the agent, disconnecting the channel, or removing access can stop that attempt. Do not submit a fresh copy while it is pending.

A dedicated Teams agent is not ready or does not reply

Open Manage agent → Channels → Microsoft Teams. An organization Owner or Admin with workspace administration access can create and manage the identity. The panel's four-step guide shows what to do next. Not connected with the app already created means the app is not installed in a team yet: select Open in Teams, complete Microsoft installation, mention the named agent in an intended standard channel, then activate that verified destination under Channel conversations in d5s. The panel checks the installation on its own while it is open; Check status checks at once.

While the panel shows Creating @Agent in Teams, d5s is still working and retries short Microsoft delays on its own. If the panel then reports Microsoft is taking longer than usual, select Retry. Nothing was lost. Microsoft rejected this step means one retry is worth it; if it fails again, contact support. Microsoft blocked this step means your Microsoft admin must allow app creation. The Teams connection expired means a workspace admin must reconnect Teams provisioning first. The small line under the message carries the technical code for support.

For thread replies without another mention, select Check status first if a connected app shows Upgrade permissions. This verifies its current permissions before you seek new consent. If the upgrade is still required, approve Upgrade permissions in Teams, return to d5s, and enable Follow replies for that channel. A grant for another team is not enough. Renew now replaces an expiring credential without asking you to copy a secret.

If an existing team installation still shows an older avatar after Sync profile, update that installation in Teams. Sync profile publishes the bot name and the agent's current description and avatar; the bot keeps its own name when you rename the agent. Microsoft Teams did not accept this name means the bot name is too long: remove the app and create it again with a name of 29 characters or fewer; a successful sync does not confirm that an existing installation has updated.

A failed management action can leave an existing connection available; retry the failed action instead of creating a duplicate identity. Removal did not finish means the identity has stopped in d5s but Microsoft cleanup is unfinished; select Retry cleanup. See Teams agent setup for the supported scope. A response in d5s is not proof that a message was sent to Teams.

Received in a Teams thread confirms that d5s accepted the request. Working means the agent has started. If no answer follows, open the agent's activity in d5s and check its outcome before sending the request again. If delivery is unknown, inspect the original thread before retrying to avoid a duplicate.

If the original Received message is confirmed but a progress update cannot be confirmed, the answer may appear as a separate ordinary Teams message. An uncertain answer send is not automatically repeated; check the thread and d5s activity before asking the agent to send it again.

After selecting Stop, wait for Stopped before treating the work as cancelled. If Teams does not show the confirmation, open the agent's activity in d5s to check the outcome. Earlier replies and completed external actions remain. Questions and approval requests keep their own controls and decision history. See Chats for the message lifecycle.

A Telegram teammate bot is missing or does not reply

If the bot is not connected, an Owner or Admin can select Add Telegram in the agent’s Channels panel and follow the illustrated setup guide. Bot connected does not mean your personal Telegram account is linked: open the bot, send /start, and confirm the d5s link before testing a reply. If the token is rejected, the setup form below the guide shows the error and lets you correct it. Use Previous and Next to revisit the instructions without clearing your token or the error, then try connecting again.

Telegram is in beta. Connecting or managing a bot requires a workspace Owner or Admin who can share and run that agent.

The agent conversation’s Channels panel shows Telegram status on one row, with an Open bot link icon when connected. Could not load Telegram means the status request failed; use the Retry icon rather than treating the bot as disconnected. For Needs attention, ask an authorized Owner or Admin to select the Telegram row and restore the connection.

If a bot will not connect, create or select it in BotFather, copy its current token again, and confirm that the bot is not already connected to another d5s agent. Each named agent needs a different bot. Never send the token in chat or a support request.

If the bot asks you to link, use a private chat with the bot, send /start, and open the returned link in a browser where you can sign in to d5s. Confirm the displayed Telegram identity with an active d5s account that still belongs to the workspace and can view and run the agent. The link expires after ten minutes; after it expires, send /start again for a fresh one. Repeated messages during the pending ceremony do not produce another link. A group member, bot account, unlinked account, or teammate whose current d5s access was removed cannot start work.

The beta accepts private text messages and one file per message: a document, photo, voice note, or audio file up to 20 MB. Groups, channels, videos, stickers, buttons, and edited messages do not wake the agent. Replies do not stream, so inspect the explicit delivery activity and the same private chat. If the agent cannot message a teammate first, check that the teammate linked their Telegram account and can still view and run the agent. If a sent file is listed as not delivered, the text still arrived; ask the agent to send the file again.

Telegram messages use a Telegram badge. If a sender's Telegram profile photo is unavailable, for example because their Telegram privacy settings hide it from bots, d5s shows their d5s avatar instead, or initials when they have none. Photo loading does not delay incoming messages. Disconnecting Telegram can make these photos unavailable.

If the outgoing activity is Accepted by Telegram, Telegram accepted the API request but has not proven teammate delivery or a read. If it is Delivery unknown, check the private chat for the original message before retrying; d5s will not blindly repeat an ambiguous send. Disconnecting the bot removes its d5s links and bindings but does not revoke its BotFather token. Revoke or rotate that token in BotFather if it must stop working. To connect again, paste a token for the same bot or for a new bot; the agent keeps one Telegram bot at a time, so disconnect the current bot before you replace it.

A Slack coworker did not receive or send a file

For an incoming file, first confirm the message shows an attachment. Ask the agent to open it when needed. Under Manage agent → Channels → Slack, follow Upgrade permissions if shown to add Slack file access. A workspace admin can reconnect an unavailable Slack setup connection there. Also confirm the agent remains installed in the channel and the file is still available in Slack. Files above the 100 MiB per-file limit are skipped with a note in the turn. One message can include any number of files.

For outgoing email, d5s-managed addresses, Gmail, and Outlook support up to 25 MiB of attachments per message. If Outlook asks you to reconnect the mailbox before it sends files of 2.5 MB or more, reconnect it once in the agent's email settings. The complete email and the recipient's service limits also apply. Reduce the attachments or send an accessible file link if delivery fails. If Gmail reports that another large attachment is being prepared, wait briefly and retry the request.

For an outgoing file, ask the agent to attach a file it created and confirm that it is available and within the attachment limits. Each file can be up to 50 MiB, and one reply can include any number of files. Missing, empty, oversized, or unsupported files are refused one by one. The chat app can also refuse a file that is above its own limit. The text reply can still arrive; inspect its attachment-failure note for the reason.

Downloaded channel files normally remain available after the coding environment stops. If a previously downloaded file is missing, give support the conversation link, message time, and visible file error. A file that was never downloaded still depends on the original remaining available in Slack.

An invitation is refused for lack of seats

For teams that purchase seats upfront, invitations cannot exceed purchased Team capacity. Active members plus pending invitations must stay within the effective seat quantity, and both organization invitations and workspace invitations count. Raise capacity from seat management or withdraw a pending invitation before trying again.

Two cases look like a full team but are not. A team organization that has never completed checkout has no Team capacity at all. Open its Settings → Organization → Billing page, select Change plan, then choose Choose seats and start Team to finish the subscription on that organization. When a paid seat decrease is already scheduled for the end of the period, new invitations are measured against the lower upcoming quantity, not today's; cancel the decrease by requesting the current quantities again if you need those seats back. Personal organizations are unaffected: their shared-workspace guests are never billable seats.

A seats-on-joining invitation needs review

If your team uses seats on joining, select Review invite and have an organization Owner or Admin authorize the displayed joining charge or free offer and individual monthly price. Quotes last up to 15 minutes. A free invitation reservation lasts up to 48 hours, bounded by the team's first billing-period end. If either expires, review and authorize new terms; the expired free offer does not silently become a paid charge.

A pending invitation is uncharged. If payment fails or billing terms change before acceptance, ask an organization Owner or Admin to authorize a new invitation. If acceptance says the purchase is being reconciled, retry as directed. Contact support if it remains blocked or requires billing reconciliation. Do not send repeated replacement invitations to bypass that state. Someone already in the team needs no additional seat and should open the workspace named in the invitation.

A message cannot be edited after compaction

A message included in a saved context summary cannot be edited. Its pencil action explains that the edit would change summarized history. Send a new message with the correction instead. Newer messages can still be edited unless replacing their conversation tail would also affect an earlier summarized input.

A seat change did not affect credits

Check the billing summary's standard and premium seat quantities and the effective date shown for the change. Increases apply after successful payment, while removals and class downgrades take effect at the end of the current billing period. If the displayed credit balance still looks wrong after the change becomes effective, contact d5s with the organization and billing period. See Plans and credits.

The result looks stale

• An existing widget board can retain its last published version during a build or failure. Open its chat and History to inspect the result.
• An app dashboard keeps its current version visible while the agent prepares a Draft. An editor reviews it and selects Update dashboard on the web, or Publish changes on mobile. On the web, find Version history in the dashboard's options menu. On mobile, select History.
• A project file or connector activated after a run changes future runs, not the history of the completed one.
• An agent keeps one long conversation and may compact older context. It can search its own summarized turns, but put durable facts in project instructions, files, or memory rather than relying on a distant chat turn.
• For widget boards, check Sources for the last successful data timestamp. A failed source refresh retains that data. For app dashboards, ask in chat to update the data, or reload the app to show data it can already read. Historical app previews use captured data and cannot fetch live results.
• Restoring a dashboard version publishes it without creating a new agent run. See Dashboards.

I cannot find a workspace setting

Type what you are looking for into Search settings at the top of the sidebar or narrow Settings index. Search matches page names and related terms: invoice finds Billing, while API keys finds Workspaces. Open Workspaces, choose a workspace, then select its Developers tab for API keys. Clear the search to return to the ordinary navigation.

Profile and Preferences are under You. Preferences combines Appearance, Sidebar shortcuts, personal Privacy, and available Labs controls. Organization administration is under Organization. People is the organization roster and invitation page. Models and Providers apply to the organization. Providers is not in the Settings list: open it from Models, or type "providers" in Search settings. To configure a workspace, open Workspaces and choose it. A plain organization member with only one accessible workspace opens its General page directly from the Workspaces shortcut. On a narrow screen, use the Settings index and the workspace and section selectors.

General is available to workspace members. Workspace Members requires an organization Owner or Admin; Network and Developers require a workspace Owner or Admin. Browser is shown where available. Search never widens those permissions. Organization Owners and Admins are also Admins of every workspace. Billing and upgrade actions require the organization role, even when you administer a workspace.

Old settings bookmarks open the corresponding current page. On a wide screen, opening Settings returns to the last saved settings page in that organization, or Profile when no such page is saved. On a narrow screen it opens the page index. Choosing another workspace to configure keeps you in Settings. It does not make that workspace the destination for closing Settings.

If General shows Workspace name as plain text instead of a field, your workspace role does not include administration. Ask a workspace Owner or Admin to rename it from Workspace settings → General → Workspace details. The name saves when they press Enter or leave the field. Organization Owners and Admins are Admins of every workspace, so they can rename it there too. The organization portfolio has no rename control. Renaming changes only the display name, so the slug and existing links remain the same.

To rename the organization itself, use the Name field in Organization > General. That page is available only to organization Owners and Admins. If it is missing, confirm that the selector under Organization in Settings shows the right organization and that your organization role is Owner or Admin. An organization rename also leaves the slug and existing links unchanged.

An agent or project cannot be deleted

Stop its active runs before deleting it. Retry deletion after the runs stop. External resource cleanup continues through the retained cleanup journal.

A workspace cannot be deleted

A workspace Owner or Admin can open Workspace settings → General → Danger zone and review what d5s will delete. Only the current Owner can request the email code and schedule deletion. Organization Owners and Admins can review and cancel as workspace Admins, but cannot schedule unless they own the workspace. Personal workspaces cannot be deleted.

The Owner can schedule deletion while content remains. They must acknowledge the impact, enter the exact workspace name, and use the email code. Workspace Admins can cancel during the 48-hour window. If ownership changes or the requester becomes inactive before cleanup starts, d5s cancels the request.

If the dialog shows Deletion needs attention, or the workspace has disappeared but the completion email has not arrived, contact support. Do not recreate the workspace to force progress.

d5s Desktop is not updating

Settings → Desktop shows the installed version at the top. Its Updates group has a manual check button and a live status line. Automatic updates download in the background and are installed when you choose Update now in Settings, select the round download button beside your avatar in the expanded sidebar, or quit the app. Before installing, d5s checks for a newer release and replaces an older downloaded update, so you can skip intermediate versions. Update now reopens d5s; a normal quit leaves it closed.

Check for Updates… in the app menu checks without opening Settings. When an update is ready, choose Install and restart to update immediately or Later to keep it queued. If no update is available, the result confirms your installed version. The sidebar download button also installs directly without opening Settings.

• macOS: "Automatic updates are unavailable in this build." This build cannot verify update signatures and cannot install an update even if one is downloaded. Download the latest installer from d5s.tech instead of waiting for it to update itself, and repeat that until this message stops appearing.
• A background check runs at launch and again every few hours; it fails silently so a temporary network problem never interrupts you with an error. Only a manual check (the button in Settings, or "Check for Updates…" in the app menu) surfaces a failure.
• Once a download finishes, a small round download button appears beside your avatar in the expanded desktop sidebar. Select it to install and reopen d5s. When the sidebar is collapsed, a small dot at the avatar's upper-right indicates the ready update but is not a button. Open Settings → Desktop → Updates to install, or quit the app. The indicator is absent while an update is still downloading and on pages without an avatar, such as sign-in. If a newer release becomes available, d5s downloads that version instead.
• If downloading a newer replacement fails in the background, Settings shows an error so you can retry; it does not report that the app is up to date.
• If d5s cannot check the latest release or finish downloading it, Update now reports a failure so you can retry. If an avatar-side install fails while that update remains ready, the expanded button shows "Update failed. Try again." and re-enables. If a newer replacement is downloading or fails, check the status in Settings and retry there. A normal quit still closes the app without installing the older queued version.

Report a bug

Select Report a bug in the sidebar, describe what went wrong, then select Send report. The report is private to the d5s team. Your account, workspace, current conversation when available, and available technical details are included automatically; you do not need to look up identifiers or device details.

A screenshot is optional. In a desktop version that supports capture, the app adds a preview of its visible content before the dialog opens. Review it and select Remove if you do not want to include it. In the browser or an older desktop version, upload or paste a screenshot instead. You can send a report without an image.

If sending fails, your description and screenshot stay in the dialog. Select Try again. After Report received, select Done to return to your work. The team can follow up at your account email address.

If you cannot open the app, contact support with a short description and the approximate time of the problem. Never include API keys, OAuth tokens, connector secrets, or confidential prompts.

If you are reporting a slow or failed API request, include its approximate time and the response's X-Request-ID header when available. This helps the team find the matching request. You can report the problem even when that header is absent.

An agent says a mailbox-channel email is still awaiting approval

Ask the agent to check the email's current outbox status. It can inspect approval and delivery records without sending another copy. A previous “queued” reply is historical; a later approval or send may already have completed. If the outcome is uncertain, avoid retrying automatically because the provider may already have accepted the message.

Agent files have not reached my Mac

Check Manage agent → Local files → Recent operations and the connection status. Keep d5s running on an awake, connected Mac. Resume paused access in Desktop settings. For an unavailable folder, select it again and review permissions. A read requires Read only or Both; a write requires Write only or Both. For an unknown write outcome, check the destination file before repeating it. Files presented in the conversation are no longer saved automatically. See Local files.

Google connection or model is unavailable

For Gmail mailbox changes, Drive file or sharing changes, Google Docs or Sheets edits, or Search Console changes, an older connection may lack Google write permission. If write access is available for your workspace, ask a workspace admin to reconnect the intended account and check Google's consent screen. If Google offers no write permission, stop and ask d5s support whether write access is available for that connection. Keep explicit Deny settings for actions you do not want. Repeated d5s approvals cannot expand Google's grant. See Connectors for supported actions and limits.

An unverified-app warning is a Google-side verification issue. A workspace setting cannot remove it. Contact d5s with the connector name and visible error, without sharing OAuth codes, tokens or authorization URLs.

If a turn was already running when Google was connected, start a new turn after setup. If the shared computer asks for a restart after Google data protection was activated, contact d5s support to restart it. This keeps its saved files. Start a new turn after the restart. If the connection or selected model is unavailable, report the visible error and selected model. Disconnecting does not erase previously retrieved information. See Google data protection for AI processing and stored-data guidance.

A Pentest VM command has not finished

Ask the agent to read the existing VM and command status. Pending startup does not mean a test has run. A queued or running command has not completed.

If a command's outcome is unknown, keep its job identifier and read that result again. Do not rerun the command to recover output.

Before stopping the VM, export any report you need to keep. Wait for confirmed cleanup before treating the VM as stopped. Optional Targets and notes are guidance and do not restrict network access.

An AWS investigation is incomplete

For a missing account, use Connect AWS account in web chat or ask a workspace admin to open Cloud accounts. For a denied operation, use Review permissions for the existing role. d5s does not change AWS policies. Role verification confirms the connection, not every service permission. Continue the read-only investigation after the admin fixes access. Scheduled read restrictions require their own review and cannot be bypassed with broader IAM permissions. See AWS recovery guidance.

A named Project request cannot continue

If a selected agent is unavailable or its eligibility could not be checked, sending is blocked. Your draft, agent mention, and files remain. Remove the agent mention to ask d5s, or type @ followed by part of another agent's name and select a match. Retry agents retries a failed lookup. An agent marked Unavailable remains visible; highlight or tap it to read the reason. You cannot select it until the issue is resolved. No matching agents means no visible agent name matches your search.

If Retry original request appears after a connection failure, use that control before sending another request in the same chat. It keeps the original text, files, and recipient. A change to an agent mention applies to later requests.

An access error can mean that the requester or a Project chat reader no longer has access to required context. Ask the workspace administrator to check Restricted Project access. Moving the chat does not grant access to its former Project. A daily budget error requires available agent budget before that agent can start more work.

A video edit or motion transfer cannot start

Check the source file named on the video card. Video editing accepts a 2–8.7 second MP4 or MOV clip. Motion transfer accepts 3–30 seconds with Follow motion video, or 3–10 seconds with Follow character image. Each video can be at most 100 MB.

Motion transfer also needs a PNG or JPEG character image of at most 10 MB. Use a clear character image and a continuous motion clip. See Context and capabilities for dimension limits.

If a source changed after the proposal, ask the agent for a new card. If the model is no longer allowed, ask an organization Owner or Admin to check Video models. A retry can incur another charge; check the card's result before retrying.