version: "1.0.1" name: use-spark description: >- Use the spark CLI to access the user's Spark email data - list emails, search by topic, read threads, check calendar events, find availability, look up contacts, and view team info. Use when the user asks about their emails, calendar, contacts, meetings, or scheduling. metadata: version: 1.2.1 requires: bins:

• spark

Using spark

spark is a CLI for the Spark email client. Use it to query the user's mailbox, calendar, contacts, meetings, and team data.

Running spark

spark <command> [options]

Environment: spark is a thin client that talks over IPC to the user's running Spark macOS Desktop app - it does not ship its own mailbox, network stack, or credentials. Run it directly on the user's Mac against the live Spark Desktop process. Do not try to execute it inside a sandbox, container, CI runner, or any environment isolated from the user's desktop session - it will fail to connect. If Spark Desktop is not running, ask the user to launch it instead of retrying.

Commands

accounts

List all configured accounts with their calendars, teams, and shared inboxes. Each account and shared inbox shows its access level in parentheses, which controls what operations Spark can perform.

spark accounts

Run this first to discover what accounts, calendars, and teams are available, and to check their access levels.

Access levels:

Access levels are configured separately for each account and each shared inbox in Spark Desktop under Settings -> AI Agents. Shared inboxes can have a different access level than the parent account - for example, a personal account may have triage access while a shared inbox under the same team is read-only or disabled.

If a command requires a higher access level than the account or shared inbox has, it returns an error with instructions on how to change the level.

folders

List folders with message counts. Output includes folder identifiers in parentheses - use these as arguments to emails and search. Mailboxes backed by a Google account show (Gmail labels) on the Email Account or Shared Inbox header. Teams show the team name as a usable identifier for emails.

spark folders                        # all accounts
spark folders user@example.com       # single account

emails

List emails with metadata (ID, From, Date, Subject, Flags). Supports pagination and Gmail-style filters.

spark emails                                                   # Unified Inbox
spark emails user@example.com:Archive                          # specific folder
spark emails "My Team"                                         # all shared threads in a team
spark emails --filter "from:alice@co.com is:unread"             # filtered
spark emails --filter "newer_than:7d has:attachment"            # recent with attachments
spark emails --page 2 --page-size 20                           # pagination
spark emails --order ascending                                 # oldest first
spark emails --new-senders                                     # show only new sender emails

GateKeeper filtering: When viewing the Inbox with GateKeeper in explicit mode, new sender emails are automatically filtered out and a "New Senders" count is shown at the top. Use --new-senders to view those emails. Use contact-action acceptContact <email> or contact-action blockContact <email> to accept or block a sender.

Folder identifier formats (run folders to see available ones):

Filter operators (combinable, Gmail-style):

search

Two modes:

• With a topic (keyword mode): Hybrid keyword + semantic search returning up to 20 emails with full bodies, sorted by relevance.
• Without a topic (list mode): Paged compact table of every email matching --filter / --in across all folders and all accounts, sorted newest first. Same output as emails, but the default scope is "all folders" instead of the Unified Inbox. Trash, Spam, and Blocked are excluded (matching Spark's search field) unless --in explicitly targets one of those folders.

spark search "quarterly report"
spark search "API integration" --filter "from:alice@co.com"
spark search "budget" --in user@example.com:Archive
spark search "vacation" --in user@example.com              # all folders in account
 
# Keywordless list mode - filter across every folder
spark search --filter "from:alice@co.com"                  # every email from alice, all folders
spark search --filter "from:alice@co.com" --in user@example.com
spark search --filter "is:unread newer_than:7d" --page 2

Use `search` with a topic when the user asks about content - it returns email bodies so you can answer questions. Use `search` without a topic when you need to filter emails (especially by `from:`) across every folder - emails only sees the Unified Inbox so it can't answer questions like "every email from alice@co.com, anywhere". Use emails for plain browsing of Inbox / one folder.

thread

Print every message in a thread - headers, full plain-text bodies, and attachment info. After the thread summary line, lists custom (non-system) folder labels once for the whole thread, using qualified names like account@domain.com:MyLabel (same style as folders).

spark thread 1114                          # by message ID from emails/search output
spark thread --download-attachments 1114   # also fetch attachments via IMAP
spark thread "https://sparkmailapp.com/dpl/bl?token=ABC..."  # by Spark deep link

The positional argument accepts either a numeric message ID (the ID: line) or a Spark deep link (the Link: line) printed by a previous run - https://sparkmailapp.com/dpl/bl?token=..., readdle-spark://bl=..., or readdlespark://bl=....

Each message's Attachments: block is a table with columns ID, Name, Size, MIME Type, and Path. The ID column is the attachment's stable pk - feed it to attachment to read the file contents (auto-downloads if necessary). The Path column shows the local file or (not downloaded, ...) for attachments not yet fetched.

Use emails or search to find message IDs (the ID column), then thread to read the full conversation. Use folders to list valid label identifiers for action attachLabel / detachLabel.

attachment

Read a single email attachment by its ID (pk) from the thread Attachments table. The file is auto-downloaded if it isn't cached locally yet.

spark attachment 42                          # print metadata (ID, Name, Size, MIME Type, Path, Message ID)
spark attachment 42 --stream > report.pdf    # write raw file bytes to stdout

Use thread to find attachment IDs (the ID column in the Attachments: table). The default text output is one Key: value per line, easy to parse from scripts.

draft

Requires: triage access level.

Create a new email draft or edit an existing one. The body is written in markdown and converted to HTML.

spark draft --to "alice@example.com" --subject "Hello" --body "Hi Alice, ..."
spark draft --to "alice@co.com" --to "bob@co.com" --cc "carol@co.com" --subject "Meeting" --body "..."
spark draft --edit 1234 --subject "Updated subject" --body "Updated body"
spark draft --reply-to 5678 --body "Thanks for the update!"
spark draft --forward 5678 --to "manager@co.com" --body "FYI"
spark draft --account "john@gmail.com" --to "alice@co.com" --subject "Hi" --body "..."
spark draft --to "alice@co.com" --subject "Report" --body "See attached" --attach /path/to/report.pdf
spark draft --to "alice@co.com" --subject "Files" --body "Two files" --attach /path/to/a.pdf --attach /path/to/b.xlsx
spark draft --to "client@co.com" --subject "Proposal" --body "..." --team "Engineering" --user alice@co.com --user bob@co.com
spark draft --edit 1234 --team "Engineering" --user alice@co.com --allow-send
spark draft --edit 1234 --user carol@co.com           # invite carol on an already-shared draft
spark draft --edit 1234 --allow-send                  # grant send-on-behalf permission on an already-shared draft
spark draft --edit 1234 --no-allow-send               # revoke previously-granted send-on-behalf permission
spark draft --edit 1234 --remove-user alice@co.com    # kick alice from a shared draft (keeps share, comments, other collaborators)
spark draft --edit 1234 --remove-user alice@co.com --user dave@co.com  # swap collaborators: remove alice, invite dave
spark draft --edit 1234 --unshare
spark draft --template "Cold outbound v3" --to "alice@co.com" --placeholder "Project name=Acme Q3" --placeholder "Deadline=Friday EOD"
spark draft --template 124 --edit 9821 --placeholder "Project name=Acme Q3" --placeholder "Deadline=Friday EOD"

Explicit flags always win over template fields. Use template <id|name> to discover the template's manual placeholders before calling draft --template - missing manual placeholders cause a hard error before any draft is created. Auto-fillable placeholders that fail to resolve (e.g. recipient name with multiple --to) leave a localized label in the body and surface in the response as a warning.

Use emails to find message IDs for --edit, --reply-to, and --forward. Use accounts to find account emails for --account - both personal accounts and shared inboxes are listed there, and either can be used as the from address when the account has draft & comment access. Use teams to find team names for --team and team member emails for --user.

Threading is critical. Whenever a new message belongs to an existing conversation, you must pass --reply-to with the last message in that thread. This is what attaches the draft to the conversation (correct In-Reply-To / References headers, same thread in the recipient's mailbox). Without --reply-to the draft starts a brand new thread, which is almost always wrong when the user asked you to "reply", "respond", "follow up", "answer", or "ping" anyone in the context of an existing conversation. Use thread <id> to inspect the conversation and pick the most recent message's ID as --reply-to.

Follow-ups (no response yet). When the user asks to follow up with someone you already emailed and they haven't replied yet (e.g. "send Alice a nudge - she never responded to my last email", "bump the proposal thread"), the most recent message in that thread is your own outgoing one. Use that message's ID as --reply-to - the follow-up stays attached to the original outgoing message so the recipient sees it as a bump on the existing conversation rather than a new cold email.

Sharing is triggered by the presence of --team or --user; teams with exactly one other active member auto-share with everyone, otherwise --user is required. To add collaborators or change the allow-send setting on an existing shared draft, use --edit <pk> together with the sharing flags - the change is applied to the existing share instead of creating a new one. To toggle allow-send off, pass --no-allow-send. To remove a specific collaborator without tearing the share down, pass --remove-user <email>; the shared draft, its comments, and the remaining collaborators stay intact. Combine --user and --remove-user in one command to swap collaborators in a single operation - removals run before invites. Content edits (--to, --cc, --bcc, --subject, --body, --attach) and sharing updates (--team, --user, --remove-user, --allow-send, --no-allow-send) must be issued as separate draft commands.

templates

List Spark message templates - the saved drafts users can apply via draft --template. Templates round-trip from desktop, so anything saved on the user's Mac shows up here.

spark templates                          # all personal + team templates
spark templates --personal               # only personal templates
spark templates --team "Marketing"       # only that team's templates
spark templates --page 2 --page-size 20  # pagination

Output columns: ID, Scope (Personal / <team name>), Name, Subject (truncated to 40 chars), Modified. Use the ID or Name value with template <id|name> and draft --template.

template

Read-only. Show a single template's full contents and its placeholder requirements. Run this before draft --template so you know which --placeholder "<name>=<value>" arguments the template needs.

spark template 123                       # by ID
spark template "Welcome reply"           # by name (case-insensitive)

Output includes scope, recipients, subject, body (HTML stripped to text), attachments, and a Placeholders: section listing every placeholder in the template:

• [auto] - auto-fillable (recipient/self name). Resolved from --to and --account when applied. Not overridable via --placeholder.
• [manual] - free-form placeholder. Required: must be passed as --placeholder "<name>=<value>" to draft --template.

If a name matches more than one template, you'll get an error listing the matching IDs - disambiguate by ID.

comment

Requires: triage access level.

Post a team comment (chat message) on a thread. If the thread is not yet shared, it will be shared automatically. Supports text comments, file attachments, or both. Use --edit to update an existing comment.

spark comment 1234 --body "Looks good, let's proceed."
spark comment 1234 --body "Please review this" --team "Engineering"
spark comment 1234 --body "FYI" --team "Engineering" --user alice@co.com --user bob@co.com
spark comment 1234 --attach /path/to/screenshot.png
spark comment 1234 --body "See attached" --attach /path/to/report.pdf --attach /path/to/data.csv
spark comment --edit 5678 --body "Updated comment text"

Use emails to find message IDs. Use thread to see comment IDs in a conversation. Use team to list teams and their members.

events

List calendar events for a time range.

spark events                                          # today's remaining events
spark events --tomorrow
spark events --week
spark events --week --in user@example.com             # specific account
spark events --week --in user@example.com:Work        # specific calendar
spark events --start 2026-03-16 --end 2026-03-20      # custom range

Date formats: yyyy-MM-dd, dd/MM/yyyy, or yyyy-MM-ddTHH:mm.

Run accounts to see available calendar accounts and calendar names.

availability

Find free time slots. Without --attendees, shows the user's own availability. With --attendees, computes mutual free windows.

spark availability                                                        # today
spark availability --tomorrow
spark availability --week --attendees alice@co.com
spark availability --start 2026-03-16 --end 2026-03-20 --attendees a@co.com,b@co.com

Free slots are within working hours (08:00-20:00), skip weekends, and ignore events marked "free".

contacts

Search contacts by name or email. Strict match first, then fuzzy fallback.

spark contacts "john"
spark contacts "example.com"

team

Show team info - metadata, shared inboxes with members, full member list, assigned emails, assignment summary.

spark team                     # list available teams
spark team "Readdle"           # specific team

meetings

List meeting transcripts with optional filters and pagination.

spark meetings
spark meetings --filter "newer_than:30d"
spark meetings --filter "subject:standup" --page-size 10

Filter operators: subject:<text>, before:yyyy/MM/dd, after:yyyy/MM/dd, newer_than:Xd, older_than:Xd.

meeting

Read a single meeting transcript's summary. Optionally include the full transcript and/or notes.

spark meeting 42                            # summary only
spark meeting --transcript 42               # include transcript
spark meeting --notes 42                    # include notes
spark meeting --transcript --notes 42       # everything
spark meeting "https://sparkmailapp.com/dpl/bl?token=ABC..."  # by Spark deep link

The positional argument accepts either a numeric meeting message ID or a Spark deep link.

Use meetings to find meeting IDs.

action

Requires: triage access level.

Perform an action on one or more emails. Supports standard email actions and team actions.

spark action <action-name> <message-id...> [options]

Supported actions:

• pin - Pin the message to keep it at the top of the list
• unpin - Remove the pin from the message
• mute - Mute the thread to stop receiving notifications
• unmute - Unmute a previously muted thread
• snooze - Snooze the message until a specific date (requires --date)
• unsnooze - Remove the snooze and return the message to the inbox
• changeReminder - Set a follow-up reminder if no reply by the date (requires --date)
• clearReminder - Remove the follow-up reminder from the message
• setAside - Set the message aside for later review
• archive - Archive the message, removing it from the inbox
• moveToInbox - Move the message back to the inbox
• moveToTrash - Move the message to trash
• moveToFolder - Move the message to a specific folder (requires --folder)
• attachLabel - Attach a Gmail or Spark Team label without removing other labels (requires --folder)
• detachLabel - Remove a Gmail or Spark Team label from the message (requires --folder)
• markAsDone - Mark the message as done
• markAsUndone - Mark the message as not done
• markAsSeen - Mark the message as read
• markAsUnseen - Mark the message as unread
• markAsSpam - Mark the message as spam
• markThreadAsPriority - Mark the thread as priority
• unmarkThreadAsPriority - Remove the priority mark from the thread
• unsubscribe - Unsubscribe from the sender or mailing list
• changeCategoryPersonal - Change the email category to Personal
• changeCategoryNotification - Change the email category to Notification
• changeCategoryNewsletters - Change the email category to Newsletter
• shareInTeam - Share the thread with teammates (requires --team when multiple teams)
• assign - Assign the email to a teammate (requires --assignee)
• delegationComplete - Mark the delegation as complete
• delegationReopen - Reopen a completed delegation

Options:

• --date - required for snooze and changeReminder, optional for assign as due date (formats: yyyy-MM-dd, dd/MM/yyyy, yyyy-MM-ddTHH:mm)
• --folder - required for moveToFolder, attachLabel, and detachLabel (qualified name: email@domain.com:FolderName; use folders to list labels including shared inboxes)
• --team - team name for team actions; required when you belong to multiple teams
• --user - teammate email to share with for shareInTeam; repeat for multiple users; required when team has more than 2 members
• --assignee - teammate email to assign the email to for assign
• --comment - comment text for the assign action

spark action pin 12345                                          # pin a message
spark action archive 100 200 300                                # archive multiple messages
spark action markAsSeen 100 200                                 # mark as read
spark action snooze 12345 --date 2026-04-10T09:00               # snooze until date/time
spark action snooze 12345 --date 2026-04-10                     # snooze until date
spark action moveToFolder 12345 --folder "user@example.com:Archive"  # move to folder
spark action attachLabel 12345 --folder "user@gmail.com:MyLabel"       # add Gmail/Team label
spark action detachLabel 12345 --folder "shared@company.com:SomeLabel"  # remove label
spark action changeReminder 12345 --date 2026-04-15             # set reminder
spark action shareInTeam 1234 --team "Engineering" --user alice@co.com  # share with teammate
spark action shareInTeam 1234 --user alice@co.com --user bob@co.com    # share with multiple
spark action assign 1234 --assignee bob@co.com                  # assign to teammate
spark action assign 1234 --assignee bob@co.com --date 2026-04-15 --comment "Please review"
spark action delegationComplete 1234                            # mark delegation done
spark action delegationComplete 100 200 300                     # complete multiple delegations
spark action delegationReopen 1234                              # reopen delegation

Use the emails command to find message IDs. Use folders to resolve qualified names for moveToFolder, attachLabel, and detachLabel. Use team to list teams and members for team actions.

contact-action

Requires: triage access level.

Perform an action on one or more contacts by email address.

spark contact-action <action-name> <email...>

Supported actions:

Examples:

spark contact-action blockContact spammer@example.com
spark contact-action acceptContact alice@co.com bob@co.com
spark contact-action changeCategoryPersonal alice@co.com
spark contact-action markContactAsPrimary ceo@company.com
spark contact-action enableAutosummaryForContact newsletter@example.com

Use the contacts command to look up email addresses.

Smart Categories

Spark automatically classifies incoming email into six categories. Use the category: filter operator with emails and search to view mail by category, and use action / contact-action to reclassify and tune.

Browse by category (read-only):

spark emails Inbox --filter "category:priority is:unread"       # unread priority mail
spark emails Inbox --filter "category:personal is:unread"       # unread people mail
spark emails Inbox --filter "category:invitation"               # pending invites
spark emails Inbox --filter "category:notification is:unread"   # unread notifications
spark emails Inbox --filter "category:newsletter newer_than:7d" # recent newsletters

Reclassify a message (triage):

spark action changeCategoryPersonal <id>       # move to People
spark action changeCategoryNotification <id>   # move to Notifications
spark action changeCategoryNewsletters <id>    # move to Newsletters

Tune per-contact category rules (triage) - changes apply to all future mail from the sender:

spark contact-action changeCategoryPersonal sender@example.com         # reclassify as People
spark contact-action changeCategoryNewsletters sender@example.com      # reclassify as Newsletters
spark contact-action groupEmailsFromContact sender@example.com         # group by category
spark contact-action markContactAsImportant vip@example.com            # enable notifications
spark contact-action markContactAsPrimary ceo@example.com              # auto-prioritize
spark contact-action enableAutosummaryForContact newsletter@example.com # AI summaries

Category-first triage pattern: Process inbox in priority order - priority first, then people, then invites, then notifications, then newsletters. This ensures the most important messages get attention first.

Typical Workflows

Answer a question about emails:

1. spark search "topic" - find relevant emails with bodies
2. Read the output and answer the user's question

Find and read a specific email:

1. spark emails --filter "from:sender subject:keyword" - locate the email
2. spark thread <ID> - read the full conversation

Draft a reply:

1. spark emails --filter "from:sender" - find the email
2. spark draft --reply-to <ID> --body "Thanks for the update!"

Send an email from a saved template:

1. spark templates - list templates (or spark templates --personal / --team "<name>")
2. spark template "<name>" - inspect placeholders before applying
3. spark draft --template "<name>" --to <recipient> --placeholder "<manual>=<value>" - create the draft (repeat --placeholder for each manual one)

Check someone's schedule for a meeting:

1. spark availability --tomorrow --attendees alice@co.com,bob@co.com
2. Suggest a time from the free slots

Comment on a shared thread:

1. spark emails --filter "from:sender" - find the email
2. spark comment <ID> --body "Looks good, approved!"

Get team workload overview:

1. spark team "Team Name" - see members and assigned emails

Look up a contact:

1. spark contacts "name or domain" - find their email address

Perform an action on emails:

1. spark emails --filter "from:sender" - find the email
2. spark action archive <ID> - archive it (or pin, snooze, etc.)

Share an email with your team:

1. spark emails --filter "subject:keyword" - find the email
2. spark action shareInTeam <ID> --user alice@co.com - share with a teammate

Delegate an email to a teammate:

1. spark emails --filter "subject:keyword" - find the email
2. spark action assign <ID> --assignee bob@co.com --date 2026-04-15 --comment "Please handle this"

Mark delegated emails as done:

1. spark emails --filter "subject:keyword" - find the email(s)
2. spark action delegationComplete <ID1> <ID2> - mark multiple as complete

Find unassigned shared inbox emails:

1. spark emails shared@co.com:Inbox --filter "assigned_to:unassigned" - list unassigned items
2. spark action assign <ID> --assignee bob@co.com - assign to a teammate

Review what's assigned to a teammate:

1. spark emails --filter "assigned_to:bob@co.com" - see their assignments
2. spark emails --filter "is:shared_inbox_open" - all open shared inbox items
3. spark emails --filter "is:shared_inbox_done" - completed items

Manage a contact:

1. spark contacts "name or domain" - find the contact email
2. spark contact-action blockContact spammer@example.com - block them (or accept, change category, etc.)

Keeping this skill up to date

The CLI and this skill share a single version (metadata.version in the front-matter above). New features in newer Spark Desktop releases ship with an updated embedded skill. Backward compatibility is guaranteed within a major version, so the commands and flags documented here keep working - but newly added features won't appear in this file until it is reinstalled.

Check for an update when - and only when - you hit one of these signals:

• The user asks about a Spark feature, command, or flag that isn't documented here.
• A spark command fails with an "unknown command" or "unknown option" error you didn't expect.
• The user explicitly says they upgraded Spark or that the skill is out of date.

In those cases:

spark --version

If the printed version is greater than metadata.version above, refresh this file before answering. spark skill prints SKILL.md to stdout - capture it and write it back over the file you're reading right now (the path is in your context). For example:

mkdir -p <parent-of-use-spark>/use-spark
spark skill > <parent-of-use-spark>/use-spark/SKILL.md

<parent-of-use-spark> is the directory containing the use-spark/ folder this SKILL.md lives in. If you have a file-write tool, you can also pipe spark skill output directly through it - the CLI itself does not write any files.

Do not check on every session or before every command - this skill is the source of truth unless one of the signals above tells you otherwise.

Tips

• Always quote multi-word arguments: spark search "project update"
• Combine filter operators in a single --filter string: --filter "from:alice@co.com is:unread newer_than:7d"
• Use folders to discover exact folder identifiers before passing them to emails or search --in
• Use accounts to discover calendar names before passing them to events --in
• The search command is best for topic-based queries; emails is best for browsing/filtering by metadata
• thread returns the full conversation - use it when you need the complete email text, not just metadata
• Use draft to compose emails - it supports new drafts, replies, forwards, and editing existing drafts
• Use comment to post team chat messages on threads - it auto-shares the thread if needed
• Use action to perform email actions like pin, archive, snooze, move to folder, and more
• Use contact-action to manage contacts - block, accept, change category, toggle auto-summary, and more