Skip to main content

MCP tool reference

Every tool the Rize MCP server exposes, with its parameters and defaults.

Read tools never modify anything. Write tools create, update, or delete records in your workspace, and clients that surface MCP annotations will ask you to confirm the destructive ones. Two tools — sign_up and get_login_url — work before you have authenticated; everything else requires a connected account.

Authenticated tool responses include a _user_timezone field. Use it to interpret and build timestamps rather than assuming UTC. Most list tools page with cursor and return has_more and next_cursor. Dashboard lists use first and after, returning page_info.has_next_page and page_info.end_cursor; follow each tool's parameter table.

note

This page is generated from the repository's registered tool definitions. Numbers, defaults, and enum values come from those schemas. The connected server's tool list is authoritative if its deployed version differs.

Session and discovery​

get_current_user​

Read

Get the authenticated user's profile including name, email, timezone, default team, team memberships (team id, name, role, status), and organization info (id, name, logo, role, finance permission). Call this first to get your org_id and team_id for other tools.

Takes no parameters.

create_checkout_url​

Write

Create a signed link to the Rize checkout page, where the user picks a plan and enters their card to subscribe their workspace. Use when a workspace admin wants to subscribe or their trial is ending, and give them checkout_url to open in their browser. Workspace admins only. Never ask for card details in chat.

Takes no parameters.

list_teams​

Read

List the teams the authenticated user can access — org admins see every team in their orgs. The user's default team is returned first. Use this to obtain a team_id for other tools (time entries, allocations, team members). Each team includes agent_context (the standing guidance applied to its members' agent runs) when the caller can edit it — team admins and org admins; null otherwise.

ParameterTypeRequiredDescription
limitnumbernoMax teams to return. Default 50.
cursorstringnoPagination cursor.

get_help​

Read

Get documentation on how to use Rize MCP tools. Pass a topic to get specific help, or omit for an overview. Topics: time_tracking, profitability, team_management, clients_projects, keywords, skills, reports_routines.

ParameterTypeRequiredDescription
topicstringnoOne of overview, time_tracking, profitability, team_management, clients_projects, keywords, skills, reports_routines. Help topic. Default "overview".

get_product_docs​

Read

Get documentation about the Rize product itself: what Rize can do, which integrations are supported, and where to find help articles. Use this to answer questions about Rize features, integrations, platforms, or setup — never guess. Pass a query to search the documentation and get the matching sections back as the docs' own prose, each with a deep link — prefer this over the static topics when the question is about a specific feature, and quote the returned steps rather than paraphrasing them. Colloquial wording works: the search widens common phrasings ("clip", "overlay", "that bar") to the product's own vocabulary. Use topic "support" whenever the docs don't cover the question, the user hit a bug, or they want to request a feature — it returns where to reach a human instead of guessing. Topics: overview, integrations, docs_index, support.

ParameterTypeRequiredDescription
topicstringnoOne of overview, integrations, docs_index, support. Documentation topic. Default "overview".
querystringnoSearch the full documentation index by keyword. Takes precedence over topic.

get_ui_map​

Read

Find where a control actually lives in Rize: the desktop menu bar, the Desktop Widget menu, and the app's settings panes, each as a full click path like "Widget > Show on Desktop". Use this whenever the user asks where a setting is, how to turn something on or off, or how to reach a screen — the paths come from the shipping navigation, so quote them exactly and never invent a screen that is not listed. An entry marked with an access note is only there for users who have that grant, so say so rather than sending everyone to it. Pair it with get_product_docs, which explains what a setting does; this tool says where to click. Pass a query to search, or omit it for the full map.

ParameterTypeRequiredDescription
querystringnoWhat the user is looking for, in their own words. Omit for the whole map.

list_skills​

Read

List the reusable prompt skills visible to the authenticated user.

Takes no parameters.

get_skill​

Read

Get one reusable prompt skill by ID. Use this when the user references a skill chip or a rize://skill/:id link.

ParameterTypeRequiredDescription
idstringyesSkill ID from a rize://skill/:id reference or list_skills result.

list_connectors​

Read

List the apps the Rize agent can use through connectors (Linear, HubSpot, Harvest, Slack, ...) and whether the user has each one connected. Call it when the user asks what is connected, or asks about an app that is not among the connected apps. Mention an app as Name: the pill opens its integration page, where an unconnected app can be connected. Tools of a connected app come from search_connector_tools.

ParameterTypeRequiredDescription
connected_onlybooleannoReturn only apps the user has connected and the agent can use now.

get_login_url​

Read · no authentication required

Returns the Rize login URL so the user can authenticate in their browser. No authentication required. Use this when the user needs to log in or connect their Rize account. When the app is needed, give the user's per-OS link from download_links following download_instructions; use downloads_url only as fallback.

Takes no parameters.

sign_up​

Write · no authentication required

Create a new Rize account. Emails the user a 6-digit sign-in code; the email has a button to the Rize sign-in page where they enter it. After signing in, the user should download the Rize desktop app to start tracking time automatically. Give the user's per-OS link from download_links following download_instructions; use downloads_url only as fallback.

ParameterTypeRequiredDescription
emailstringyesUser's email address.
namestringnoUser's full name.
timezonestringnoIANA timezone (e.g. America/New_York). Defaults to UTC.
app_sign_inbooleannoWhen true, the emailed code deep-links into the installed Rize desktop app (rize://) instead of the browser sign-in page. Use only after the user has installed the desktop app; defaults to false for users who have not.

get_setup_checklist​

Read

Get the optional getting-started checklist the Rize app shows after onboarding: the user's persona, its steps, and which are done. This is not onboarding. Onboarding is finished when save_onboarding_answers returns onboarded true, so never use open steps here to say the user isn't onboarded. Don't list the steps unprompted; when the user asks what to do next, offer the single most useful open step, and do it yourself when you can. Electron_installed tells you whether the desktop app is tracking yet. When the app is needed, give the user's per-OS link from download_links following download_instructions; use downloads_url only as fallback.

Takes no parameters.

save_onboarding_answers​

Write

Save the user's Rize sign-up answers, the same ones the web getting-started flow asks, and report whether onboarding is finished. Use it for "am I onboarded?" and "finish my onboarding". Fill in every answer you already know from context (memory, earlier conversation, connected tools) before asking, and pass only answers you have evidence for. Call it with no answers to read missing_answers. Onboarding completes on its own once name, use_case, company_name and firm_type are saved (name and use_case for members of someone else's workspace): the trial starts and the web app stops showing getting-started. missing_answers lists the required answers still unsaved; ask the user only for those you can't work out. The other answers are the rest of that flow. Save each of them too, so the app has nothing left to ask.

ParameterTypeRequiredDescription
namestringnoThe user's full name.
use_casestringnoOne of individual, team. individual for "For myself", team for "With my team".
company_namestringnoThe company or workspace name.
firm_typestringnoOne of agency, consultancy, accounting, law, other. How the user describes their company: Agency, Accounting firm, Consultancy, Law firm, or other with firm_type_other.
firm_type_otherstringnoThe firm type in the user's words, when firm_type is other.
job_titlestringnoThe user's role as the sign-up label, e.g. "Owner / Founder / CEO" or "Project / Account Manager".
goalsstring[]noAny of actuals, profitability, timesheets, budgets, capacity, focus. What the user wants from Rize: actuals = how long our work really takes, profitability = which clients we're losing money on, timesheets = stop chasing timesheets, budgets = projects going over budget, capacity = how much more work the team can take on, focus = focus and productivity.
goals_otherstringnoThe goal in the user's words, when they pick "Something else".
track_targetsstring[]noAny of clients, projects, tasks. What the team tracks time against. Pass every level they use: the workspace keeps those on and switches the others off.
sign_up_sourcestringnoHow the user heard about Rize: referral (From a friend), team_invite (From my team), claude, chatgpt, perplexity, gemini, google (Google Search), twitter, clickup, linkedin, instagram, windows_app_store, youtube, tiktok, newsletter, reddit, zenpilot, or their own answer in snake_case.
silent_modebooleannotrue for Silent (tracks quietly, no notifications), false for Productivity (focus and break nudges).
timezonestringnoThe user's IANA time zone, e.g. America/New_York.
twenty_four_hour_timebooleannotrue for 24-Hour times, false for AM/PM.

Time entries​

list_my_time_entries​

Read

List the current user's own time entries for a date range. For team-wide entries (admin only), use list_team_time_entries instead. Returns all statuses by default (tracking, segmenting, active, pending, generating, failed); rejected entries are never returned, so you can see live tracked and unprocessed time as well as reviewable suggestions. Sorted by start time. Each entry is a compact summary: id, times, duration, title, description, status, source, billable, client/project/task/label ids and names, pending tag suggestions, and confidence (0-100, how sure Rize is, on new-tracker workspaces only); empty fields are omitted. Each entry also has start_time_local/end_time_local in the user's timezone, and the response has totals (overall and by status, partial: true when has_more) — quote these instead of adding durations or converting UTC yourself. For billing amounts, rates, AI reasoning, team, creator, or tag-suggestion reasoning, call get_time_entry for one entry or pass include_details: true. Returns 50 entries by default; when has_more is true, pass next_cursor as cursor to get the next page.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date for single-day query).
querystringnoSearch time entries by title or description.
statusesstring[]noAny of tracking, segmenting, active, pending, generating, failed. Array of time entry statuses to include, for example ["active", "pending"]. "tracking" is time currently being tracked live, "segmenting"/"generating" are being processed into suggestions, "pending"/"failed" are suggestions awaiting review. Do not pass a single string.
client_idsstring[]noFilter by client IDs.
project_idsstring[]noFilter by project IDs.
task_idsstring[]noFilter by task IDs.
sourcesstring[]noAny of ai, ai_edited, system, user. Filter by who authored the entry description and duration: "ai" = fully AI-generated and unedited, "ai_edited" = AI-generated then edited by the user, "user" = created by a person, "system" = written without AI because the workspace was out of AI credits. This is what the Rize app labels "Source".
creation_sourcesstring[]noFilter by the internal creation channel, not returned in results, for example ["ai", "click_up", "user", "timer", "meeting"]. Useful for finding duplicates or verifying sync.
label_idsstring[]noFilter by label IDs.
team_idstringnoFilter by team ID (omit to include all of your teams).
include_activitybooleannoInclude top_apps and top_titles, plus at most 20 keyword-match windows per entry, grouped by keyword/target within this page (counts and seconds are page-local). keyword_matches_total_count counts all raw windows; keyword_matches_truncated marks partial evidence. Call get_time_entry with include_keyword_matches and keyword_matches_offset=keyword_matches_next_offset to page remaining raw windows. Slower — only set when tagging or inspecting entries.
include_detailsbooleannoReturn the full field set per entry: billable amount, hourly rate, currency, AI reasoning, team, creator, and tag-suggestion reasoning. Use get_time_entry instead when only a few entries need detail.
limitnumbernoMax entries to return. Default 50.
cursorstringnoPagination cursor from previous response.

Scoped to the authenticated user regardless of role. Use list_team_time_entries for other people's time.

list_team_time_entries​

Read

List time entries across all team members (team admin only). Returns entries for the entire team by default. Use creator_emails to filter to specific people. Non-admins will only see their own entries. Sorted by start time. Each entry is a compact summary: id, times, duration, title, description, status, source, billable, client/project/task/label ids and names, creator id/name/email, and pending tag suggestions; empty fields are omitted. Each entry also has start_time_local/end_time_local in the user's timezone, and the response has totals (overall and by status, partial: true when has_more) — quote these instead of adding durations or converting UTC yourself. For billing amounts, rates, AI reasoning, team, or tag-suggestion reasoning, call get_time_entry for one entry or pass include_details: true. For hours by person, client, or project, use get_team_time_allocation instead of paging entries. Returns 50 entries by default (max 200); when has_more is true, pass next_cursor as cursor to get the next page. Prefer paging with next_cursor over raising limit.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date for single-day query).
querystringnoSearch time entries by title or description.
creator_emailsstring[]noFilter to specific team members by email (e.g. ['macgill@rize.io']).
creator_idsstring[]noFilter to specific team members by identity ID (the number in a rize://member/<id> pill; alternative to creator_emails).
statusesstring[]noAny of tracking, segmenting, active, pending, generating, failed. Array of time entry statuses to include, for example ["active", "pending"]. "tracking" is time currently being tracked live, "segmenting"/"generating" are being processed into suggestions, "pending"/"failed" are suggestions awaiting review. Do not pass a single string.
client_idsstring[]noFilter by client IDs.
project_idsstring[]noFilter by project IDs.
task_idsstring[]noFilter by task IDs.
sourcesstring[]noAny of ai, ai_edited, system, user. Filter by who authored the entry description and duration: "ai" = fully AI-generated and unedited, "ai_edited" = AI-generated then edited by the user, "user" = created by a person, "system" = written without AI because the workspace was out of AI credits. This is what the Rize app labels "Source".
creation_sourcesstring[]noFilter by the internal creation channel, not returned in results, for example ["ai", "click_up", "user", "timer", "meeting"]. Useful for finding duplicates or verifying sync.
label_idsstring[]noFilter by label IDs.
team_idstringnoFilter by team ID (omit to include every team you can view).
include_detailsbooleannoReturn the full field set per entry: billable amount, hourly rate, currency, AI reasoning, team, and tag-suggestion reasoning. Use get_time_entry instead when only a few entries need detail.
limitnumbernoMax entries to return. Default 50.
cursorstringnoPagination cursor from previous response.

get_time_entry​

Read

Get a single time entry by ID with all details including client, project, task, billing info, and AI confidence data. Set include_tag_explanation to see why the entry is tagged the way it is: which keyword rule or AI suggestion set each tag, and on new-tracker workspaces Rize's reasoning for the time range, tags, and label with a confidence per decision (the top-level confidence field is the weakest of those decisions). Use it before fixing a wrong or missing tag or when the user asks why an entry was tagged a certain way. Set include_screen_vision to also get the Vision screen captures taken during the entry. Set include_keyword_matches for raw keyword-match windows (at most 20), total count and truncated flag. Page with keyword_matches_offset=keyword_matches_next_offset until it is null.

ParameterTypeRequiredDescription
idstringyesThe time entry ID.
include_keyword_matchesbooleannoReturn a page of raw keyword-match windows for this entry.
keyword_matches_offsetnumbernoOffset from keyword_matches_next_offset; defaults to 0.
keyword_matches_limitnumbernoRaw windows per page, default/max 20.
include_tag_explanationbooleannoAlso explain the entry's tags: which keyword rule or AI suggestion set the client, project and task, with evidence, plus Rize's step-by-step reasoning (time range, client, project, task, label) with confidence and what decided each on new-tracker workspaces. Use before fixing a wrong or missing tag or when asked why an entry was tagged a certain way.
include_screen_visionbooleannoAlso return the Vision screen captures taken during this entry, oldest first, 50 per page. Each capture is a timestamp, the app or website, a short summary of what was on screen, and a screenshot link when the workspace stores screenshots. Links expire after 15 minutes; describe the summary rather than opening images unless the user asks to see one. Empty when Vision is off for the account. Use when the user asks what they were actually doing or looking at during an entry.
vision_cursorstringnoWith include_screen_vision, pass vision_captures_next_cursor from the previous response to get the next page of captures.

create_time_entry​

Write

Create a time entry with optional client, project, and task assignment. If a matching active entry overlaps the requested range, Rize extends and updates that entry instead. Supports idempotency keys to prevent duplicate entries on retry. Times must be in ISO 8601 format — convert user-local times to their timezone (provided as _user_timezone in responses) before sending.

ParameterTypeRequiredDescription
start_timestringyesStart time in ISO 8601 format (e.g. 2024-01-15T09:00:00Z).
end_timestringyesEnd time in ISO 8601 format.
titlestringnoTitle for the time entry.
descriptionstringnoDescription of work performed.
client_idstringnoClient ID to assign.
project_idstringnoProject ID to assign.
task_idstringnoTask ID to assign.
team_idstringnoTeam ID (defaults to user's default workspace).
billablebooleannoWhether this entry is billable.
idempotency_keystringnoUnique key to prevent duplicate entries on retry. Recommended for all creates.

batch_create_time_entries​

Write

Batch create 2+ entries for a timesheet, recurring blocks, or per-day hours. For one entry use create_time_entry. Pass local dates and times; they are converted using your timezone. Convert percentage splits to hours of the day total. Log PTO or holidays on an existing workspace project, found with list_projects; never invent one. Dry-run first if times are not exact or overlaps may exist. Report created and failed counts from the result; never assume success.

ParameterTypeRequiredDescription
entriesobject[]yesRows with a local date and either start/end times or hours; gap rows require hours.
day_startstringnoLocal start time for the first hours row on each date. Default "09:00".
team_idstringnoWorkspace team to associate with each entry.
idempotency_keystringnoBase key; the tool appends each row's index.
dry_runbooleannoPreview rows and overlaps without creating entries. Default false.

update_time_entry​

Write

Update an existing time entry. Supports changing times, title, description, billing, label, and entity reassignment (client, project, task). Changing team_id clears entity assignments. To retag many entries with the same client, project, task, or label, use batch_update_time_entries instead of calling this per entry.

ParameterTypeRequiredDescription
idstringyesThe time entry ID to update.
start_timestringnoNew start time in ISO 8601 format.
end_timestringnoNew end time in ISO 8601 format.
titlestringnoNew title.
descriptionstringnoNew description.
client_idstringnoReassign to this client, or null to clear.
project_idstringnoReassign to this project, or null to clear.
task_idstringnoReassign to this task, or null to clear.
label_idstringnoAssign this label to the time entry, or null to clear.
team_idstringnoMove to this team (clears entity assignments).
billablebooleannoSet billable status.

batch_update_time_entries​

Write

Batch update matching active or pending time entries with one set of client, project, task, label, or billable changes. For one entry use update_time_entry. Use it instead of repeated update_time_entry calls when one change applies to several entries. Dry-run first, show the user the count and sample, and get confirmation before calling with dry_run: false, unless you pass entry_ids for 20 or fewer entries the user asked to change. Times, titles, team assignments, and clearing fields are not supported.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date for a single-day range).
entry_idsstring[]noOnly update these entries, using IDs from list_my_time_entries. The date range and other filters still apply, so pass the range the entries are in.
querystringnoSearch time entries by title or description.
client_idsstring[]noFilter by client IDs.
project_idsstring[]noFilter by project IDs.
task_idsstring[]noFilter by task IDs.
label_idsstring[]noFilter by label IDs.
sourcesstring[]noAny of ai, ai_edited, system, user. Filter by time entry source.
team_idstringnoFilter by team ID.
statusesstring[]noAny of active, pending. Array of statuses to include; tracking and generating entries are excluded. Default ["active","pending"].
creator_emailsstring[]noFilter by team member email; requires appropriate team permissions.
creator_idsstring[]noFilter by team member identity ID; requires appropriate team permissions.
client_idstringnoAssign this client to matching entries.
project_idstringnoAssign this project to matching entries.
task_idstringnoAssign this task to matching entries.
label_idstringnoAssign this label to matching entries.
billablebooleannoSet billable status on matching entries.
dry_runbooleannoPreview changes without saving. Default true.

split_time_entry​

Write

Split a time entry into two at split_at. The second part keeps the original's tags unless you pass overrides (e.g. a different client). Use when one entry covers two clients or tasks. Approved and pending entries can be split; a pending suggestion splits into two pending halves.

ParameterTypeRequiredDescription
idstringyesThe time entry ID to split.
split_atstringyesSplit point as an ISO 8601 datetime with Z or an explicit offset.
titlestringnoOverride the second part's title.
client_idstringnoOverride the second part's client.
project_idstringnoOverride the second part's project.
task_idstringnoOverride the second part's task.
billablebooleannoOverride the second part's billable status.

merge_time_entries​

Write

Merge two or more of your time entries into one. The earliest entry is kept and stretched to span all of them, keeping its tags; the others are deleted. Use when Rize cut one piece of work into several entries. Pending suggestions stay pending.

ParameterTypeRequiredDescription
idsstring[]yesIDs of the time entries to merge (at least two).

delete_time_entry​

Write

Delete a time entry by ID. Works on entries of any status (active, pending, failed, etc.). Only call it after the user has explicitly confirmed this deletion.

ParameterTypeRequiredDescription
idstringyesThe time entry ID to delete.

batch_delete_time_entries​

Write

Batch delete several time entries by ID in one call. For one entry use delete_time_entry. Runs a dry run by default; show the user the count and entries and get confirmation before calling again with dry_run false.

ParameterTypeRequiredDescription
entry_idsstring[]yesThe IDs of the time entries to delete.
dry_runbooleannoPreview the entries without deleting them. Default true.

Live sessions​

get_current_session​

Read

Return the user's running focus, meeting, or break session, or running: false when none is active. Check this before start_session or stop_session.

Takes no parameters.

start_session​

Write

Start a live focus, meeting, or break session timer for the user, the same as the Start button in the Rize app. Starting a session ends any session that is already running.

ParameterTypeRequiredDescription
typestringnoOne of focus, meeting, break. Session type. Default "focus".
minutesnumbernoPlanned length in minutes. Omit for an open-ended session.
titlestringnoWhat the user is working on.
client_idstringnoClient ID to tag the session with.
project_idstringnoProject ID to tag the session with.
task_idstringnoTask ID to tag the session with.

stop_session​

Write

Stop the user's running focus, meeting, or break session, the same as the Stop button in the Rize app. Returns stopped: false when no session is running.

Takes no parameters.

AI suggestions and notes​

generate_time_entries​

Write

Generate AI time entries for a time range. Analyzes the user's actual activity — apps, websites, meetings — and uses clustering to create multiple entries based on natural activity groups. By default, skips time slots where previous entries were rejected. Rate limited: 15 per minute.

ParameterTypeRequiredDescription
start_timestringyesStart time in ISO 8601 format (e.g. 2024-01-15T00:00:00-04:00).
end_timestringyesEnd time in ISO 8601 format (e.g. 2024-01-15T23:59:59-04:00).
team_idstringnoTeam ID (defaults to user's default workspace).
retry_rejectedbooleannoIf true, also generates entries in time slots where previous entries were rejected. Default false.

Generation is asynchronous. The call returns immediately; poll list_my_time_entries with statuses: ["generating", "pending"] until the entries settle. Shares a quota with regenerate_time_entry: 15 calls per minute and 100 per day.

regenerate_time_entry​

Write

Regenerate AI content for a pending or failed time entry. Useful when generation failed or you want a better title/description. Optionally provide custom instructions to guide the AI. Rate limited: max 3 regenerations per entry, 15 per minute.

ParameterTypeRequiredDescription
idstringyesThe time entry ID to regenerate.
instructionsstringnoCustom instructions to guide the AI regeneration (e.g. 'focus on the meeting with Client X').

Regeneration is asynchronous and capped at 3 attempts per entry — the 4th call returns an error. Shares the generation quota above.

approve_time_entries​

Write

Approve pending AI-generated time entry suggestions, making them active entries. Optionally assign client/project/task during approval in a single operation. Returns total_duration_seconds and total_duration_formatted; quote the formatted total rather than adding durations.

ParameterTypeRequiredDescription
idsstring[]yesArray of time entry IDs to approve.
client_idstringnoAssign this client to all approved entries.
project_idstringnoAssign this project to all approved entries.
task_idstringnoAssign this task to all approved entries.

reject_time_entries​

Write

Reject pending AI-generated time entry suggestions. Rejected entries are kept but hidden from active views.

ParameterTypeRequiredDescription
idsstring[]yesArray of time entry IDs to reject.

approve_tag_suggestion​

Write

Approve an AI-generated tag suggestion (client, project, or task) on a time entry. This assigns the suggested entity to the time entry. Use list_my_time_entries to see tag suggestions with confidence scores on pending entries.

ParameterTypeRequiredDescription
idstringyesThe tag suggestion ID to approve.

add_note​

Write

Add a note about what you're working on. Notes give Rize context to improve time tracking accuracy.

ParameterTypeRequiredDescription
textstringyesWhat did you work on? Natural language.
datestringnoReference date YYYY-MM-DD. Defaults to today.
savebooleannoSet true to save time entries after previewing. Only relevant when blocks are provided. Default false.
billablebooleannoOverride billable status for created entries.
blocksobject[]noPre-parsed time blocks. When provided with durations > 0, time entries will be created in addition to the note. Tag each block with any combination of client, project, and/or task.

Each entry in blocks is an object: description (string, required), durationMin (number, 0–1440, required), and optional client, project, and task names matched against your existing entities. A block with durationMin: 0 records context without creating an entry. Blocks are laid out backwards from the anchor time, so list them in the order the work happened.

With blocks and save: false (the default) the tool returns a preview — the entries it would create, the client/project/task it matched each block to, and any overlap with entries that already exist. Call again with save: true to commit.

dictate​

Write

DEPRECATED: Use add_note instead. This tool now delegates to add_note.

ParameterTypeRequiredDescription
textstringyesNatural language time description.
datestringnoReference date YYYY-MM-DD. Defaults to today.
savebooleannoSet true to save after previewing. Defaults to false (preview). Default false.
billablebooleannoOverride billable status.
blocksobject[]noPre-parsed blocks. Tag each block with any combination of client, project, and/or task.

Deprecated. It delegates to add_note and takes the same arguments — call add_note directly.

Raw activity​

list_my_apps_used​

Read

List the authenticated user's own apps and websites used in a date range, sorted by time spent. Returns app name, URL, time spent, and category.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date).

list_my_events​

Read

List raw tracking events (app switches, website visits) for the authenticated user in a date range. Max 31 calendar days. Returns app name, URL, URL host, title, source, and timestamps. Use list_my_apps_used for aggregated summaries instead.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date for single-day query).
limitnumbernoMax events per page. Default 200.
cursorstringnoPagination cursor from previous response.

The date range is capped at 31 calendar days. For an aggregated view, use list_my_apps_used.

list_my_vision_captures​

Read

List the authenticated user's Vision screen captures in a date range, oldest first. Max 7 calendar days. Each capture is a timestamp, the app or website, a short summary of what was on screen, and a screenshot link when the workspace stores screenshots. Links expire after 15 minutes; describe the summary rather than opening images unless the user asks to see one. Empty when Vision is off for the account. For the captures inside one time entry, call get_time_entry with include_screen_vision instead.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date for single-day query).
limitnumbernoMax captures per page. Default 100.
cursorstringnoPagination cursor from previous response.

list_my_calendar_events​

Read

List calendar events (meetings, appointments) for a single day from the user's connected calendars. Returns title, start/end times, attendees, location, and video conference link per event. For raw tracked activity (app switches, website visits) use list_my_events instead.

ParameterTypeRequiredDescription
datestringyesDate in YYYY-MM-DD format.

search_my_meetings​

Read

Search the current user's calendar events and meeting transcripts over a date range of up to 372 calendar days. Use this for what was scheduled, who attended, what was discussed, or finding the recording behind a meeting.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format; defaults to date.
querystringyesWords or phrase to find in calendar details or transcript content.
limitnumbernoMaximum transcript matches. Default 50.

list_my_keyword_matches​

Read

List deterministic keyword-rule matches over the user's tracked time in a date range. By default, returns totals per keyword and target; pass detail: true for individual match windows. Use these as ground truth when tagging or creating time entries.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date for single-day query).
detailbooleannoReturn individual match windows instead of summaries. Default false.
limitnumbernoMax windows in detail mode. Default 200.

Analytics​

get_my_time_allocation​

Read

Get the current user's own time allocation summary. By default returns the full view — breakdowns by client, project, task, AND label — plus totals and billable hours. Pass group_by to fetch a single breakdown, including title. For team-wide allocation (admin only), use get_team_time_allocation instead.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date).
group_bystringnoOne of client, project, task, label, title. Single breakdown to return. Omit for the full view (client + project + task + label breakdowns).
client_idsstring[]noFilter by client IDs.
project_idsstring[]noFilter by project IDs.
task_idsstring[]noFilter by task IDs.
label_idsstring[]noFilter by label IDs.
team_idstringnoFilter by team ID (omit to include all of your teams).
member_titlesstring[]noFilter by workspace member titles.
member_title_nullbooleannoInclude entries whose workspace member title is empty.
statusesstring[]noAny of tracking, segmenting, active, pending, generating, failed. Array of time entry statuses to include, for example ["active"]. "tracking" is time currently being tracked live, "segmenting"/"generating" are being processed into suggestions, "pending"/"failed" are suggestions awaiting review. Do not pass a single string. Default ["active"].

get_team_time_allocation​

Read

Get time allocation summary across all team members (team admin only). Returns total hours, billable hours, and breakdown by grouping, including workspace member title. Use creator_emails or member_titles to filter to specific people. Filter by sources to split manual (user) vs AI time. Non-admins will only see their own allocation. With team_id and group_by member, every active team member who tracks time is returned, with total 0 if they logged nothing; use this to find who hasn't logged time.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date).
creator_emailsstring[]noFilter to specific team members by email (e.g. ['macgill@rize.io']).
creator_idsstring[]noFilter to specific team members by identity ID (the number in a rize://member/<id> pill; alternative to creator_emails).
group_bystringnoOne of client, project, task, label, member, title. How to group results. Default "client".
bucketstringnoOne of none, day. Optional daily breakdown: "day" splits each group per calendar day. Default "none".
client_idsstring[]noFilter by client IDs.
project_idsstring[]noFilter by project IDs.
task_idsstring[]noFilter by task IDs.
label_idsstring[]noFilter by label IDs.
team_idstringnoFilter by team ID (omit to include every team you can view).
member_titlesstring[]noFilter by workspace member titles.
member_title_nullbooleannoInclude entries whose workspace member title is empty.
statusesstring[]noAny of tracking, segmenting, active, pending, generating, failed. Array of time entry statuses to include, for example ["active"]. "tracking" is time currently being tracked live, "segmenting"/"generating" are being processed into suggestions, "pending"/"failed" are suggestions awaiting review. Do not pass a single string. Default ["active"].
sourcesstring[]noAny of ai, ai_edited, system, user. Filter by who authored the entry description and duration: "ai" = fully AI-generated and unedited, "ai_edited" = AI-generated then edited by the user, "user" = created by a person, "system" = written without AI because the workspace was out of AI credits. This is what the Rize app labels "Source".
creation_sourcesstring[]noFilter by the internal creation channel, not returned in results, for example ["ai", "click_up", "user", "timer", "meeting"]. Useful for finding duplicates or verifying sync.

query_time_data​

Read

Answer ad hoc time questions the fixed allocation tools can't: break hours, billable hours and entry counts down by one or two dimensions (e.g. member then project), bucket by day, week or month, and compare with the prior period of equal length. Scoped by the caller's team role: admins and viewers see every member, managers see their direct reports, members see only their own time. Returns at most 500 rows.

ParameterTypeRequiredDescription
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date).
group_bystringnoOne of client, project, task, label, member, title. First breakdown. Default "member".
then_bystringnoOne of client, project, task, label, member, title. Optional second breakdown inside each of the top 10 first-level groups, e.g. group_by member then_by project.
bucketstringnoOne of none, day, week, month. Split each row by day, week (Monday start) or calendar month. Default "none".
compare_to_priorbooleannoAlso return totals and per-group change versus the prior period of the same length. Default false.
team_idstringnoFilter by team ID (omit to include every team you can view).
creator_emailsstring[]noFilter to specific team members by email.
client_idsstring[]noFilter by client IDs.
project_idsstring[]noFilter by project IDs.
task_idsstring[]noFilter by task IDs.
label_idsstring[]noFilter by label IDs.
statusesstring[]noAny of active, pending, generating, failed. Array of time entry statuses to include, for example ["active"]. Default ["active"].
sourcesstring[]noAny of ai, ai_edited, system, user. Filter by entry source: "ai", "ai_edited", "system" or "user".

get_workspace_time_allocation​

Read

Get time allocation across every team and person in the workspace in one call (workspace admins only; pass workspace_id from get_current_user org.id). Supports member and title grouping plus per-day buckets — ideal for workspace reports and timesheet compliance (compact member×day summaries). Day buckets use the caller's timezone. Returns total hours, billable hours, and breakdown by grouping. Filter by sources to split manual (user) vs AI time. With group_by 'member' and no client/project/task/label filter, every accepted workspace member is returned — members with no entries matching the requested statuses (default: active) appear with total_duration_seconds 0 and no date, historical groups may also include deactivated or removed members. For accepted-member headcounts, paginate list_workspace_members to completion, filter status to accepted and apply the same team/title/creator filters, then match distinct identity_id values to group id. Preserve historical allocation totals.

ParameterTypeRequiredDescription
workspace_idstringyesWorkspace ID. Get it from get_current_user (org.id). You must be a workspace admin.
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date).
group_bystringnoOne of client, project, task, label, member, title. How to group results. Default "member".
bucketstringnoOne of none, day. Optional daily breakdown: "day" splits each group per calendar day in your timezone. Default "none".
creator_emailsstring[]noFilter to specific team members by email (e.g. ['macgill@rize.io']).
creator_idsstring[]noFilter to specific team members by identity ID (the number in a rize://member/<id> pill; alternative to creator_emails).
client_idsstring[]noFilter by client IDs.
project_idsstring[]noFilter by project IDs.
task_idsstring[]noFilter by task IDs.
label_idsstring[]noFilter by label IDs.
team_idsstring[]noFilter by team IDs.
member_titlesstring[]noFilter by workspace member titles.
member_title_nullbooleannoInclude entries whose workspace member title is empty.
statusesstring[]noAny of active, pending, generating, failed. Array of time entry statuses to include, for example ["active"]. Do not pass a single string. Default ["active"].
sourcesstring[]noAny of ai, ai_edited, system, user. Filter by who authored the entry description and duration: "ai" = fully AI-generated and unedited, "ai_edited" = AI-generated then edited by the user, "user" = created by a person, "system" = written without AI because the workspace was out of AI credits. This is what the Rize app labels "Source".
creation_sourcesstring[]noFilter by the internal creation channel, not returned in results, for example ["ai", "click_up", "user", "timer", "meeting"]. Useful for finding duplicates or verifying sync.

get_ai_effectiveness_stats​

Read

Get AI effectiveness metrics for time entry creation and tagging. Shows acceptance rates, tagging coverage, label coverage, AI vs manual breakdown, confidence scores, and trends. Use to assess how well Rize's AI is performing for a team or user.

ParameterTypeRequiredDescription
start_timestringyesStart of date range (ISO 8601 datetime, e.g. 2025-01-01T00:00:00Z).
end_timestringyesEnd of date range (ISO 8601 datetime, e.g. 2025-06-01T00:00:00Z).
team_idstringnoTeam ID to filter by. Defaults to the user's default workspace.
creator_idsstring[]noFilter by specific identity IDs (the number in a rize://member/<id> pill).
creator_emailsstring[]noFilter by creator emails (resolved to identity IDs).
exclude_identity_idsstring[]noIdentity IDs to exclude from results (e.g. admin accounts).
include_weekly_trendbooleannoInclude week-over-week trend data. Default false.
include_member_breakdownbooleannoInclude per-member stats breakdown. Default false.
week_countnumbernoNumber of weeks for trend data (max 52). Default 12.

get_my_time_tracking_signals​

Read

Get your recent time tracking signals — the individual AI actions and user feedback events that make up your time tracking history. Returns events like entry accepts/rejects, tag suggestions, label changes, and manual edits with full metadata (reasoning, confidence, tag_type, before/after values). Use alongside get_ai_effectiveness_stats to drill into the 'why' behind aggregate metrics.

ParameterTypeRequiredDescription
event_typesstring[]noFilter by event types. Available: time_entry_accepted, time_entry_rejected, time_entry_deleted, time_entry_auto_approved, time_entry_regenerated, time_entry_split, time_entry_merged, time_entry_created_manually, title_changed, description_changed, tag_suggestion_accepted, tag_suggestion_rejected, tag_suggestion_auto_approved, manual_tag_change, batch_tag_applied, time_entry_created, tag_suggestion_created, tag_rule_created, tag_rule_removed.
time_entry_idstringnoFilter signals for a specific time entry.
start_timestringnoStart of date range (ISO 8601 datetime). Defaults to 30 days ago.
end_timestringnoEnd of date range (ISO 8601 datetime). Defaults to now.
limitnumbernoNumber of signals to return (max 50). Default 25.
offsetnumbernoOffset for pagination. Default 0.

Clients, projects, and tasks​

list_clients​

Read

List clients (customers/accounts) with their hourly rates and team associations. Use client IDs when creating or updating time entries.

ParameterTypeRequiredDescription
querystringnoSearch clients by name.
queriesstring[]noSearch clients by multiple names. Cannot be combined with query; cursor is ignored and next_cursor is null.
statusesstring[]noArray of client statuses to include, for example ["active"]. Do not pass a single string. Default ["active"].
limitnumbernoMax clients to return. Default 50.
cursorstringnoPagination cursor.
include_keywordsbooleannoInclude keywords (auto-tagging rules) for each client. Off by default for performance. Default false.

create_client​

Write

Create a client (customer/account), or update and reactivate an existing client with the same name. Clients are top-level entities that projects and time entries can be assigned to.

ParameterTypeRequiredDescription
namestringyesClient name.
hourly_ratenumbernoDefault hourly rate for billing.
colorstringnoColor hex code (e.g. #FF5733).
team_namestringnoTeam name to associate with (defaults to user's default team). Creates the team if no team with this name exists. To add an existing client to another team, use update_client with add_team_names.

update_client​

Write

Update an existing client's name, hourly rate, color, status, keywords (for auto-tagging keyword matching), or prompt (AI context for tagging). Share the client with more teams (add_team_names) or stop sharing it (remove_team_names) so those teams can track time to the same client.

ParameterTypeRequiredDescription
idstringyesThe client ID to update.
namestringnoNew name.
hourly_ratenumbernoNew hourly rate.
colorstringnoNew color hex code.
statusstringnoNew status (e.g. active, archived).
keywordsstring[]noKeywords for auto-tagging. When the AI sees these strings in window titles, URLs, or app names, it matches the time entry to this client. Use specific terms like client names, domains, or project codes.
promptstringnoAI context prompt for tagging. Describes the client so the AI can better match time entries. Example: 'Acme Corp — SaaS company. Google Ads campaigns, Shopify store management, Klaviyo email flows.'.
add_team_namesstring[]noExact team names to share this client with. Each team gets the same org client with its rate and settings. Requires team admin on each team, or org admin. Already-shared teams are left as is.
remove_team_namesstring[]noExact team names to stop sharing this client with. Existing time entries are kept.

list_projects​

Read

List projects with their client associations and team info. Use project IDs when creating or updating time entries.

ParameterTypeRequiredDescription
querystringnoSearch projects by name.
queriesstring[]noSearch projects by multiple names. Cannot be combined with query; cursor is ignored and next_cursor is null.
client_idstringnoFilter by client ID.
statusesstring[]noArray of project statuses to include, for example ["in_progress", "completed"]. Do not pass a single string. Default ["in_progress","completed"].
limitnumbernoMax projects to return. Default 50.
cursorstringnoPagination cursor.
include_keywordsbooleannoInclude keywords (auto-tagging rules) for each project. Off by default for performance. Default false.

create_project​

Write

Create a project, or update an existing project with the same name, optionally under a client. Projects organize time entries and can be assigned to time entries directly.

ParameterTypeRequiredDescription
namestringyesProject name.
client_idstringnoClient ID to associate with.
client_namestringnoClient name to associate with (creates client if it doesn't exist).
colorstringnoColor hex code.
team_namestringnoTeam name to associate with (defaults to user's default team).

update_project​

Write

Update an existing project's name, client, color, status, keywords (for auto-tagging keyword matching), or prompt (AI context for tagging).

ParameterTypeRequiredDescription
idstringyesThe project ID to update.
namestringnoNew name.
client_idstringnoClient ID to associate with.
client_namestringnoClient name to associate with (creates if doesn't exist).
colorstringnoNew color hex code.
statusstringnoNew status (e.g. in_progress, completed, archived).
keywordsstring[]noKeywords for auto-tagging. When the AI sees these strings in window titles, URLs, or app names, it matches the time entry to this project. Use specific terms like repo names, Jira codes, or unique identifiers.
promptstringnoAI context prompt for tagging. Describes the project so the AI can better match time entries. Example: 'Website Redesign — Next.js migration. Work in Figma, VS Code with repo acme/website-v2, Vercel deploys.'.

list_tasks​

Read

List tasks with their project and assignee associations. Use task IDs when creating or updating time entries.

ParameterTypeRequiredDescription
querystringnoSearch tasks by name.
queriesstring[]noSearch tasks by multiple names. Cannot be combined with query; cursor is ignored and next_cursor is null.
project_idsstring[]noFilter tasks by project IDs.
assigned_to_mebooleannoOnly return tasks assigned to the current user.
statusesstring[]noArray of task statuses to include, for example ["in_progress", "completed"]. Do not pass a single string. Default ["in_progress","completed"].
limitnumbernoMax tasks to return. Default 50.
cursorstringnoPagination cursor.
include_keywordsbooleannoInclude keywords (auto-tagging rules) for each task. Off by default for performance. Default false.

create_task​

Write

Create a task, or update an existing task with the same name, optionally under a project or with a due date/reminder. Tasks are the most granular unit of work and can be assigned to team members.

ParameterTypeRequiredDescription
namestringyesTask name.
project_idstringnoProject ID to associate with.
project_namestringnoProject name to associate with (creates project if it doesn't exist).
assignee_emailstringnoEmail of team member to assign (defaults to authenticated user).
colorstringnoColor hex code.
team_namestringnoTeam name to associate with (defaults to user's default team).
due_datestringnoISO 8601 due date for the task (e.g. 2026-08-12).
reminder_atstringnoISO 8601 datetime for a one-time reminder (e.g. 2026-08-12T16:30:00-04:00). Use the user's timezone.

update_task​

Write

Update an existing task's name, project, assignee, color, status, or keywords (for auto-tagging keyword matching).

ParameterTypeRequiredDescription
idstringyesThe task ID to update.
namestringnoNew name.
project_idstringnoProject ID to associate with.
project_namestringnoProject name to associate with (creates if doesn't exist).
assignee_emailstringnoEmail of team member to assign.
colorstringnoNew color hex code.
statusstringnoNew status (e.g. in_progress, completed).
keywordsstring[]noKeywords for auto-tagging. When the AI sees these strings in window titles, URLs, or app names, it matches the time entry to this task. Use specific terms like ticket IDs, branch names, or unique identifiers.

Labels​

list_labels​

Read

List labels available for tagging time entries. Labels with Auto-apply off remain assignable explicitly; use label IDs when updating time entries.

ParameterTypeRequiredDescription
querystringnoSearch labels by name.
statusesstring[]noArray of label statuses to include, for example ["active"]. Do not pass a single string. Default ["active"].
limitnumbernoMax labels to return. Default 50.
cursorstringnoPagination cursor.

create_label​

Write

Create a label for categorizing time entries. Requires team admin role. Auto-apply is on by default; disabled labels remain assignable explicitly.

ParameterTypeRequiredDescription
namestringyesLabel name.
descriptionstringyesHuman-readable description of what this label represents.
promptstringnoAI prompt for automatic classification. Required when auto_label_enabled is true or omitted.
counts_toward_work_hoursbooleannoWhether time with this label counts toward the member's work-hours total.
auto_label_enabledbooleannoWhether AI may apply this label automatically (default true; at most 30 per team). Labels with this off can still be assigned explicitly.
colorstringnoColor hex code (e.g. #FF5733).
team_idstringnoTeam ID to associate with.
team_namestringnoTeam name to associate with (defaults to user's default team).

apply_workspace_template​

Write

Apply a starter label set for a firm type (accounting, agency, consultancy, law) to a team, plus how to structure clients and projects. Dry run by default; confirm before applying. Creates missing labels and reactivates archived or deleted ones without changing their settings; active labels are skipped.

ParameterTypeRequiredDescription
team_idstringyesTeam ID to apply the template to.
templatestringyesOne of accounting, agency, consultancy, law, default. Firm type label template to apply.
dry_runbooleannoPreview labels without creating or restoring them. Default true.

update_label​

Write

Update a label's name, description, prompt, color, status, or Auto-apply setting. Requires team admin role. Disabled labels remain assignable explicitly.

ParameterTypeRequiredDescription
idstringyesThe label ID to update.
namestringnoNew name.
descriptionstringnoNew description.
promptstringnoAI prompt for automatic classification. Required when auto_label_enabled is true or omitted.
counts_toward_work_hoursbooleannoWhether time with this label counts toward the member's work-hours total.
auto_label_enabledbooleannoWhether AI may apply this label automatically (default true; at most 30 per team). Labels with this off can still be assigned explicitly.
colorstringnoNew color hex code.
statusstringnoNew status (active or archived).

delete_label​

Write

Delete a label by ID. Requires team admin role. The label is soft-deleted and will no longer appear in label lists.

ParameterTypeRequiredDescription
idstringyesThe label ID to delete.

Keywords and tagging settings​

list_keywords​

Read

List active keywords (auto-tagging rules) for the current user. Keywords map text patterns to clients, projects, or tasks — when a keyword appears in a window title, URL, or app name, the time entry is auto-tagged to the parent entity. Must specify tag_type to scope the query.

ParameterTypeRequiredDescription
tag_typestringyesOne of client, project, task. Required: filter by entity type — client, project, or task.
tag_idstringnoFilter by specific client/project/task ID.

create_keyword​

Write

Create a new keyword (auto-tagging rule). Keywords auto-tag time entries when the keyword text matches in window titles, URLs, or app names. Each keyword maps to a client, project, or task.

ParameterTypeRequiredDescription
keywordstringyesThe keyword to match (e.g. 'acme', 'acme.com', 'jira.atlassian.net').
tag_typestringyesOne of client, project, task. The entity type to tag to: client, project, or task.
tag_idstringyesThe ID of the client, project, or task to tag to.
match_typestringnoOne of contains, starts_with, ends_with, equals. How to match: contains (default), starts_with, ends_with, or equals. Default "contains".
fieldstringnoOne of any, app_name, url, window_title, calendar_event. Which field to match against: any (default), app_name, url, window_title, or calendar_event. Default "any".

update_keyword​

Write

Update an existing keyword's text, match type, or field. Use list_keywords to find the keyword ID first.

ParameterTypeRequiredDescription
idstringyesThe keyword ID to update.
keywordstringnoNew keyword text.
match_typestringnoOne of contains, starts_with, ends_with, equals. New match type.
fieldstringnoOne of any, app_name, url, window_title, calendar_event. New field to match against.

delete_keyword​

Write

Delete (archive) a keyword. The keyword will no longer be used for auto-tagging. Use list_keywords to find the keyword ID first.

ParameterTypeRequiredDescription
idstringyesThe keyword ID to delete.

get_tagging_settings​

Read

Get the current user's auto-tagging and work-hours settings: daily work-hours target per weekday (hours), generation mode, entry duration preferences, auto-approve threshold, which tag dimensions are automated, the custom instructions for tagging and activity summaries, and the agent guidance layered onto their runs (personal, plus their default team/workspace guidance). Read-only — use update_tagging_settings to change your own settings. Admins: manage other members via get/update_member_agent_settings, teams via update_team_agent_context, and the workspace via update_workspace_agent_context.

Takes no parameters.

update_tagging_settings​

Write

Update your AI tagging settings: tracking mode, minimum entry duration, auto-approve threshold, and custom instructions for how the AI should tag your time entries and generate activity summaries. Which tag dimensions are automated is user-managed and not changeable here. Also sets your daily work-hours target per weekday. Use get_tagging_settings to read current values first.

ParameterTypeRequiredDescription
tracking_generation_modestringnoOne of tag_rules_and_clustering, tag_rules_only, clustering. How time entries are matched to tags. 'tag_rules_and_clustering' (recommended) combines keyword matching with AI. 'clustering' uses AI only. 'tag_rules_only' uses keywords only.
keyword_matching_enabledbooleannoTurns keyword rules on or off for tracked time. Off switches tracking_generation_mode to 'clustering' (AI only); on switches it to 'tag_rules_and_clustering'.
minimum_time_entry_minutesnumbernoMinimum duration in minutes for a time entry (default 8, recommend 2-5 for detailed tracking).
auto_approve_thresholdnumbernoConfidence threshold (0-100) for auto-approving AI tag suggestions. Default 95 (very strict). Recommend 70-85 for most users.
custom_instructions_for_taggingstringnoCustom instructions for how the AI should tag your time entries to clients/projects/tasks. Describe your workflow, main clients, and how to identify which client from window titles. Example: 'I manage Google Ads for 3 clients. Match by ad account name in browser title.'.
custom_instructions_for_activity_summarystringnoCustom instructions for how the AI should generate activity summaries (time entry titles/descriptions). Example: 'Use task names from ClickUp. Keep titles under 60 chars. Include client name prefix.'.
agent_contextstringnoPersonal guidance injected into every agent run (chat, reports, routines, tagging). May embed skills as markdown links like Name; they are expanded at run time. Team/org guidance is admin-managed and read-only here.
work_hours_per_dayobjectnoDaily work-hours target in hours (0-24) for the weekdays passed, e.g. {"monday": 7.5}. Omitted days stay unchanged.

update_team_agent_context​

Write

Set a team's standing agent guidance — injected into every team member's agent runs (chat, reports, routines, tagging). Team admins and org admins only. Read current values via list_teams. May embed skills as markdown links like Name; only team- or workspace-visible skills are accepted. Pass an empty string to clear.

ParameterTypeRequiredDescription
team_idstringyesThe team to update (from list_teams).
agent_contextstringyesThe new team guidance. Empty string clears it.

update_workspace_agent_context​

Write

Set the workspace's standing agent guidance — injected into every member's agent runs (chat, reports, routines, tagging). Workspace admins only. Read the current value via get_tagging_settings (org_agent_context). May embed skills as markdown links like Name; only workspace-visible skills are accepted. Pass an empty string to clear.

ParameterTypeRequiredDescription
agent_contextstringyesThe new workspace guidance. Empty string clears it.
org_idstringnoThe workspace to update. Defaults to the caller's active workspace (see get_current_user).

Contracts and profitability​

list_contracts​

Read

List contracts for an organization. Contracts track billing arrangements with clients including hourly rates, retainers, and profitability metrics. Archived contracts are excluded by default — pass status to filter. Use contract IDs with get_contract_profitability. Use this for contract terms like retainer hours, fixed fees, and monthly budgets.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
querystringnoSearch contracts by name.
statusstringnoOne of draft, active, paused, completed, archived. Filter by contract status.
limitnumbernoMax contracts to return. Default 50.
cursorstringnoPagination cursor.

get_contract​

Read

Get a single contract with all its periods and profitability details. movable_time is time still on archived contracts for dates this contract covers; move_time_to_contract moves it here.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
idstringyesContract ID.

create_contract​

Write

Create a new contract for profitability tracking. Contracts define billing arrangements (hourly, retainer, fixed fee) with clients. Automatically creates the first contract period. Use get_current_user to get org_id. Pass client_name or org_client_id to link a client.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
namestringyesContract name.
contract_typestringnoOne of client, internal, pro_bono. Contract type. Default "client".
billing_modelstringnoOne of hourly, retainer, hybrid, fixed_fee, internal, custom. Billing model. Default "hourly".
billing_periodstringnoOne of monthly, quarterly, yearly, one_time. Billing period. Default "monthly".
hourly_ratenumbernoHourly rate.
retainer_amount_centsnumbernoRetainer amount in cents.
hours_includednumbernoHours included in retainer.
overage_hourly_ratenumbernoOverage hourly rate.
currencystringnoCurrency code. Default "USD".
notesstringnoContract notes.
org_client_idstringnoOrganization client ID to link.
client_namestringnoClient name to look up (alternative to org_client_id).
period_start_datestringnoFirst period start date (ISO 8601, defaults to start of current month).
period_end_datestringnoFirst period end date (ISO 8601, defaults to the end of the billing interval covering today; quarters and years run from the start date).

update_contract​

Write

Update a contract's billing details. Changes to rate fields are synced to the current period.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
idstringyesContract ID to update.
namestringnoContract name.
contract_typestringnoOne of client, internal, pro_bono. Contract type.
billing_modelstringnoOne of hourly, retainer, hybrid, fixed_fee, internal, custom. Billing model.
statusstringnoOne of draft, active, paused, completed, archived. Contract status.
billing_periodstringnoOne of monthly, quarterly, yearly, one_time. Billing period.
hourly_ratenumbernoHourly rate.
retainer_amount_centsnumbernoRetainer amount in cents.
hours_includednumbernoHours included in retainer.
overage_hourly_ratenumbernoOverage hourly rate.
currencystringnoCurrency code.
notesstringnoContract notes.
org_client_idstringnoOrganization client ID to link.

move_time_to_contract​

Write

Move time that is still on archived contracts onto this contract, for the dates this contract's periods cover. Use after replacing a client's contract when get_contract shows movable_time. Never create copies of time entries to fix contract totals.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
idstringyesContract ID that should receive the time.

get_org_profitability​

Read

Get aggregated profitability metrics across all non-archived contracts for an organization in a date range. Returns revenue, costs, margin, and hours. For per-contract detail use get_contract_profitability. All monetary values are in cents.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
start_datestringyesStart date (ISO 8601, e.g. 2025-01-01).
end_datestringyesEnd date (ISO 8601, e.g. 2025-01-31).

Requires org admin or the finance permission. All monetary values are in cents.

get_contract_profitability​

Read

Get profitability metrics for a specific contract in a date range. Returns revenue, costs, margin, hours, budget burn, and period dates. Use list_contracts to find contract IDs. All monetary values are in cents.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
idstringyesContract ID.
start_datestringyesStart date (ISO 8601, e.g. 2025-01-01).
end_datestringyesEnd date (ISO 8601, e.g. 2025-01-31).

get_profitability_trend​

Read

Get monthly revenue, cost, and expense totals for a date range. Returns one data point per month across all non-archived contracts. Useful for spotting trends and comparing periods. All monetary values are in cents.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
start_datestringyesStart date (ISO 8601, e.g. 2025-01-01).
end_datestringyesEnd date (ISO 8601, e.g. 2025-06-30).

create_expense​

Write

Add an expense to a contract period. Expenses can be pass-through, delivery, or overhead. Categories: ad_spend, vendor, freelancer, software, other. Get the contract_period_id from get_contract. Metrics recompute automatically after adding.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
contract_period_idstringyesContract period ID to add the expense to.
expense_typestringyesOne of pass_through, delivery, overhead. Type: pass_through (billed to client), delivery (cost of delivery), overhead.
amount_centsnumberyesAmount in cents.
datestringyesExpense date (ISO 8601, must fall within the contract period).
categorystringyesOne of ad_spend, vendor, freelancer, software, other. Expense category.
currencystringnoCurrency code. Default "USD".
descriptionstringnoDescription of the expense.
vendor_namestringnoVendor name.

create_revenue_entry​

Write

Add a revenue entry to a contract period. Categories: setup_fee, consulting, upsell, adjustment, other. Get the contract_period_id from get_contract.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
contract_period_idstringyesContract period ID to add revenue to.
amount_centsnumberyesAmount in cents.
datestringyesRevenue date (ISO 8601, must fall within the contract period).
categorystringyesOne of setup_fee, consulting, upsell, adjustment, other. Revenue category.
currencystringnoCurrency code. Default "USD".
descriptionstringnoDescription of the revenue entry.

Invoices​

list_invoices​

Read

List client invoices for an organization, newest issue date first, with status, amounts, balance and due date. Filter by status (e.g. overdue or sent for unpaid invoices), client, contract, or a search on invoice number or client name. Org admins only. Use get_current_user to get org_id and get_invoice for line items and payments.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
statusesstring[]noAny of draft, sent, viewed, partially_paid, paid, overdue, void. Array of statuses; only invoices in these statuses.
workspace_client_idsstring[]noOnly invoices for these clients (workspace_client_id from list_clients).
contract_idstringnoOnly invoices for this contract.
querystringnoSearch by invoice number or client name.
limitnumbernoMax invoices to return. Default 50.
cursorstringnoPagination cursor.

get_invoice​

Read

Get one client invoice with its line items, recorded payments, billing details and activity history (created, sent, viewed, payments). Amounts are in cents.

ParameterTypeRequiredDescription
idstringyesInvoice ID.

list_uninvoiced_time_entries​

Read

List a client's billable time entries that are not on any live invoice yet, newest first, with duration, rate and billable amount. Use before create_invoice to pick time_entry_ids, or to report unbilled time per client. Org admins on plans with invoicing only.

ParameterTypeRequiredDescription
workspace_client_idstringyesClient workspace_client_id (from list_clients).
contract_idstringnoOnly time attributed to this client contract.
start_datestringnoEarliest entry date (YYYY-MM-DD).
end_datestringnoLatest entry date (YYYY-MM-DD).
limitnumbernoMax entries to return. Default 200.
cursorstringnoPagination cursor.

create_invoice​

Write

Create a draft client invoice. It is not sent; send_invoice emails it or marks it as sent. Bill time by passing time_entry_ids from list_uninvoiced_time_entries (lines are priced from the contract), pass manual line_items, or pass only a retainer or fixed-fee contract_id to bill its terms. Pass workspace_client_id or client_name. To copy an invoice issued in another system such as Xero, use invoice_type freeform and pass its number, issue_date, due_date, currency and line_items that add up to its total, with tax as its own line; then call send_invoice with send_email false, and record_invoice_payment for what was paid. Org admins on plans with invoicing only.

ParameterTypeRequiredDescription
org_idstringyesOrganization (billing) ID.
workspace_client_idstringnoClient workspace_client_id (from list_clients).
client_namestringnoExact client name (alternative to workspace_client_id).
contract_idstringnoContract to bill under; must belong to the client.
invoice_typestringnoOne of tracked, freeform, retainer, fixed_fee, hybrid. tracked bills time entries; freeform is manual lines only. Default "tracked".
numberstringnoYour own invoice number, e.g. INV-4905 from an accounting system; must be unused. Omit for the next Rize number.
issue_datestringyesIssue date (YYYY-MM-DD).
payment_termsstringnoOne of upon_receipt, net_15, net_30, net_45, net_60. Sets the due date from the issue date; ignored when due_date is given. Default "net_30".
due_datestringnoDue date (YYYY-MM-DD), set directly instead of from payment_terms.
currencystringnoThree-letter ISO 4217 code, e.g. AUD. Defaults to the contract's or client's currency.
time_entry_idsstring[]noUninvoiced time entries to bill.
groupingstringnoOne of task, project, member, detailed. How time entries roll up into lines (default task).
line_itemsobject[]noManual lines; when given, time_entry_ids is ignored.
subjectstringnoInvoice subject, e.g. 'October 2026 services'.
notesstringnoNotes shown on the invoice.
recipientsstring[]noEmails send_invoice will deliver to.

update_invoice​

Write

Edit a draft invoice: subject, notes, dates, payment terms, billing details or recipients. add_line_items appends lines and line_items replaces every line, in the same save as the other changes. Only recipients can change once an invoice is sent.

ParameterTypeRequiredDescription
idstringyesInvoice ID.
subjectstringno
notesstringno
payment_instructionsstringno
billing_contact_namestringno
billing_addressstringno
issue_datestringnoIssue date (YYYY-MM-DD).
payment_termsstringnoOne of upon_receipt, net_15, net_30, net_45, net_60.
due_datestringnoDue date (YYYY-MM-DD), set directly instead of from payment_terms.
recipientsstring[]noEmails send_invoice will deliver to.
line_itemsobject[]noReplaces every line item.
add_line_itemsobject[]noLines to append.

send_invoice​

Write

Email an invoice to its recipients and mark a draft as sent; calling it on a sent or overdue invoice resends it. This emails the client, so confirm with the user first. With send_email false it only marks a draft as sent and emails nobody: use that for an invoice the client already received, such as one copied from an accounting system.

ParameterTypeRequiredDescription
idstringyesInvoice ID.
send_emailbooleannofalse marks a draft as sent without emailing anyone. Default true.
sent_datestringnoDay the invoice was sent (YYYY-MM-DD), only with send_email false; use the issue date when the source gives no sent date. Defaults to today.

record_invoice_payment​

Write

Record a payment received outside Rize against a sent invoice. The invoice becomes partially_paid or paid, and a paid invoice takes its paid date from payment_date. The amount cannot exceed the remaining balance. This emails nobody.

ParameterTypeRequiredDescription
invoice_idstringyesInvoice ID.
amount_centsnumberyesAmount received in cents.
payment_datestringyesDate received (YYYY-MM-DD). If the source shows the invoice as paid without a date, ask the user for it.
payment_methodstringyesOne of bank_transfer, check, credit_card, cash, other. How it was paid.
notesstringnoe.g. a check number or transfer reference.

void_invoice​

Write

Void an unpaid invoice. Its time entries become uninvoiced again. Recipients get a void notice by email if Rize emailed them the invoice; send_email overrides that either way. Paid invoices cannot be voided.

ParameterTypeRequiredDescription
idstringyesInvoice ID.
send_emailbooleannotrue emails recipients a void notice, false emails nobody. Default: email only if Rize emailed them the invoice.

Teams and members​

list_workspace_members​

Read

List the workspace (organization) roster with workspace titles, departments, and per-team assignments. Visibility depends on your role: workspace admins see everyone, team admins and viewers see their teams, managers see their direct reports, plain members see an empty roster. Rates are only returned for workspace admins and finance grantees. Get workspace_id from get_current_user (org.id). Pass identity_ids to look up specific members by identity ID.

ParameterTypeRequiredDescription
workspace_idstringyesWorkspace (organization/billing) ID.
querystringnoSearch members by name, email, or title.
identity_idsstring[]noFilter to specific members by identity ID (the number in a rize://member/<id> pill). Use this instead of query when you already have the ID.
limitnumbernoMax members to return. Default 50.
cursorstringnoPagination cursor.

list_team_members​

Read

List team members with their workspace-level title and department, team roles, managers, hourly rates, and cost rates. Any team member can list: org and team admins and viewers see the whole roster, managers see their direct reports plus themselves, members see only themselves. Archived members are never returned. Rates require team admin, org admin, or finance permissions. Cost rates affect profitability calculations (delivery_labor_cost_cents). Pass identity_ids to look up specific members by identity ID. Pass include_tracking_status to add each member's last desktop tracking time and platform (last 30 days; null for members you can only see as yourself), e.g. to find who has stopped tracking.

ParameterTypeRequiredDescription
team_idstringyesTeam ID to list members for.
querystringnoSearch members by name or email.
identity_idsstring[]noFilter to specific members by identity ID (the number in a rize://member/<id> pill). Use this instead of query when you already have the ID.
statusesstring[]noArray of team member statuses to include, for example ["active"]. Do not pass a single string. Default ["active"].
include_tracking_statusbooleannoAdd last_tracked_at and last_tracked_platform for each member. Default false.
limitnumbernoMax members to return. Default 50.
cursorstringnoPagination cursor.

get_member_agent_settings​

Read

Get another workspace member's agent settings: their personal guidance, tagging instructions, and activity summary instructions. Admins only (org admins, plus admins of an active team the member belongs to) — returns not-found otherwise. Find identity ids via list_workspace_members. For your OWN settings use get_tagging_settings.

ParameterTypeRequiredDescription
identity_idstringyesThe member's identity id (from list_workspace_members).
org_idstringnoThe workspace the member belongs to. Defaults to the caller's active workspace.

update_member_agent_settings​

Write

Update another workspace member's agent settings: their personal guidance, tagging instructions, and/or activity summary instructions. Admins only (org admins, plus admins of an active team the member belongs to). Omitted fields are left unchanged; pass an empty string to clear one. Read current values first with get_member_agent_settings. For your OWN settings use update_tagging_settings.

ParameterTypeRequiredDescription
identity_idstringyesThe member's identity id (from list_workspace_members).
org_idstringnoThe workspace the member belongs to. Defaults to the caller's active workspace.
agent_contextstringnoThe member's personal agent guidance. Empty string clears it.
custom_instructions_for_taggingstringnoThe member's personal tagging instructions. Empty string clears it.
custom_instructions_for_activity_summarystringnoThe member's activity summary instructions. Empty string clears it.

invite_team_member​

Write

Invite a new member to a team by email. New members get an invitation email; re-inviting an existing member returns the existing record without sending another email. Acceptance links are delivered only to the invitee by email, not returned to the inviter. Requires team admin or org admin permissions. Title is stored on the workspace member, so setting it requires org admin access.

ParameterTypeRequiredDescription
team_idstringyesTeam ID to invite the member to.
emailstringyesEmail address of the person to invite.
namestringnoName of the person being invited.
titlestringnoWorkspace member title. Setting it requires org admin access.
rolestringnoOne of member, manager, admin, viewer. Role to assign (default: member). Managers see direct reports only; viewers have read-only visibility and are excluded from tracking rosters. Default "member".
billable_by_defaultbooleannoWhether the member's time is billable by default (default: true).
manager_idstringnoTeam member ID of this member's manager (must be an admin or manager on the same team).

invite_team_members​

Write

Invite up to 25 people to one team in a single call. New members get an invitation email; people already on the team get no new email. Acceptance links are delivered only to the invitee by email, not returned to the inviter. Each result has requested_email (the address passed in) and, on success, the member record. Same permissions as invite_team_member. One failed invite does not stop the rest; check each result's error.

ParameterTypeRequiredDescription
team_idstringyesTeam ID to invite the members to.
invitesobject[]yesPeople to invite. Each needs an email; name, role, title, billable_by_default, and manager_id are optional.

invite_workspace_member​

Write

Invite a member to a workspace (organization) without adding them to a team. Sends an invitation email. Requires workspace admin; Rize employees may target any workspace via organization_id.

ParameterTypeRequiredDescription
organization_idstringnoWorkspace/organization ID to invite into. Omit to invite into the caller's own workspace.
emailstringyesEmail address of the person to invite.
namestringnoName of the person being invited.
titlestringnoWorkspace member title.
rolestringnoOne of member, admin. Workspace role to assign. Default "member".

update_team_member​

Write

Update a team member's team role, manager, hourly rate, cost rate, billable default, or workspace-level title. Team fields require team admin or org admin permissions. Title updates require self or org admin access. Use list_team_members to find team_member_id values.

ParameterTypeRequiredDescription
team_idstringyesTeam ID.
team_member_idstringyesTeam member ID to update.
rolestringnoOne of member, manager, admin, viewer. Role to assign. Managers see direct reports only; viewers have read-only visibility and are excluded from tracking rosters.
titlestringnoWorkspace member title. Updating another member's title requires org admin access.
manager_idstringnoTeam member ID of this member's manager (must be an admin or manager on the same team). Pass null to clear.
hourly_ratenumbernoHourly billing rate.
cost_ratenumbernoInternal cost rate per hour.
billable_by_defaultbooleannoWhether time is billable by default.

remove_team_member​

Write

Remove a member from a team. The membership is archived, not deleted — historical time entries are kept and re-inviting the person restores it. Requires team admin or org admin permissions. Use list_team_members to find team_member_id values.

ParameterTypeRequiredDescription
team_idstringyesTeam ID.
team_member_idstringyesTeam member ID to remove.

Reports, dashboards, and routines​

list_report_runs​

Read

List report run summaries for the current user's reports, ordered by most recent first. Use get_report_run for full analyses. Filter by report ID or status (pending, running, ready, failed).

ParameterTypeRequiredDescription
report_idstringnoFilter runs to a specific report ID.
statusstringnoOne of pending, running, ready, failed. Filter by run status.
limitnumbernoMax runs to return. Default 25.
cursorstringnoPagination cursor from previous response.

export_grouped_time_entries​

Write

Email the caller a zip with one CSV of a team's time entries per group (member, client, project, task, label, department, title, or day). Team admins only. The export runs in the background and arrives by email.

ParameterTypeRequiredDescription
team_idstringyesTeam ID.
datestringyesStart date in YYYY-MM-DD format.
end_datestringnoEnd date in YYYY-MM-DD format (defaults to date).
group_bystringyesOne of member, client, project, task, label, department, title, day. One CSV file per value of this dimension.
team_member_idsstring[]noLimit to these team member IDs.
client_idsstring[]no
project_idsstring[]no
task_idsstring[]no
label_idsstring[]no
include_billable_amountbooleannoDefault true.

get_report_run​

Read

Get a single report run by ID, including the parent report metadata and all AI analyses with summaries and insights.

ParameterTypeRequiredDescription
idstringyesReport run ID.

run_report​

Write

Run a report now. report_id is the report's public ID from list_report_runs. Team and org reports run for every member and use AI credits, so confirm with the user first. Returns the pending run; check it with get_report_run. The run sends its output to the report's configured email or Slack destinations.

ParameterTypeRequiredDescription
report_idstringyesThe report's public ID from list_report_runs (report.public_id).
range_start_datestringnoOptional start date for a one-off custom range (YYYY-MM-DD).
range_end_datestringnoOptional end date for a one-off custom range (YYYY-MM-DD).

create_report​

Write

Create a report: a recurring summary of tracked time (sections plus optional AI analysis) delivered in-app, by email, or in Slack. Start from a template_key or set sections and a prompt; other fields override the template. New reports are drafts and never run on schedule until published with update_report status active (or create with status active when the user asked to turn it on).

ParameterTypeRequiredDescription
template_keystringnoTemplate key by report scope. Individual: daily_summary (daily wrap-up), personal_weekly_review (weekly trends and top work), personal_monthly_review (monthly trends and shifts), productivity_report (focus quality and time use), focus_report (focus vs meetings and breaks), meeting_review (meeting load and recurring meetings), apps_websites_report (screen time by app/site). Team: daily_team_visibility (team time by member), weekly_team_overview (team hours, top work and trends), monthly_team_review (workload, capacity and focus shifts), client_report (team hours by client), project_report (time by project), task_report (team hours by task), labels_report (effort by label), team_hours_report (hours per member), meeting_overview (team meeting load). Org: client_profitability_report (client revenue, cost and margin), multi_team_overview (hours across teams). Scratch: scratch (build your own).
namestringnoReport name.
promptstringnoWhat the AI analysis should cover each run.
sectionsstring[]noAny of work_hours, summary, ai_analysis, focus, meetings, apps_websites, clients, projects, tasks, labels, teams, members, client_profitability. Sections to include, replacing the current list.
output_formatstringnoOne of structured, ai_generated. structured (sections) or ai_generated (one AI write-up).
cadencestringnoOne of daily, weekday, weekly, monthly. Schedule cadence.
schedule_enabledbooleannoWhether the report runs on its schedule once active. false = on demand.
schedule_time_of_daystringnoLocal time HH:MM, e.g. 08:00.
schedule_day_of_weeknumbernoWeekly cadence: 0 = Sunday, 1 = Monday ... 6 = Saturday.
schedule_day_of_monthZodUnionnoMonthly cadence: 1-28, or -1 for the last day.
range_daysnumbernoReport on the last N days instead of the previous period; null returns to the previous period.
delivery_channelsstring[]noAny of in_app, email, slack. Where each run is delivered.
email_recipientsstring[]noExtra email addresses (personal reports only).
recipient_audiencestringnoOne of creator, team_admins, team, org_admins, org. Team and org reports: who receives each run.
connector_slugsstring[]noExternal connectors the AI analysis may use (e.g. notion, linear).
scopestringnoOne of individual, team, org. Whose data it reads (default individual).
statusstringnoOne of draft, active. draft (default) or active.
runs_forstringnoOne of personal, team, org. Who it runs for (default personal). Team and org need admin rights.
team_idstringnoTeam ID when runs_for is team.
org_idstringnoWorkspace ID; defaults to the user's default workspace.

update_report​

Write

Edit a report's name, sections, AI prompt, schedule, date range, delivery, or connectors. Omitted fields stay unchanged. Set status active to publish a draft, paused to stop its schedule, or draft to unpublish. Team and org reports need admin rights.

ParameterTypeRequiredDescription
report_idstringyesThe report's public ID.
namestringnoReport name.
promptstringnoWhat the AI analysis should cover each run.
sectionsstring[]noAny of work_hours, summary, ai_analysis, focus, meetings, apps_websites, clients, projects, tasks, labels, teams, members, client_profitability. Sections to include, replacing the current list.
output_formatstringnoOne of structured, ai_generated. structured (sections) or ai_generated (one AI write-up).
cadencestringnoOne of daily, weekday, weekly, monthly. Schedule cadence.
schedule_enabledbooleannoWhether the report runs on its schedule once active. false = on demand.
schedule_time_of_daystringnoLocal time HH:MM, e.g. 08:00.
schedule_day_of_weeknumbernoWeekly cadence: 0 = Sunday, 1 = Monday ... 6 = Saturday.
schedule_day_of_monthZodUnionnoMonthly cadence: 1-28, or -1 for the last day.
range_daysnumbernoReport on the last N days instead of the previous period; null returns to the previous period.
delivery_channelsstring[]noAny of in_app, email, slack. Where each run is delivered.
email_recipientsstring[]noExtra email addresses (personal reports only).
recipient_audiencestringnoOne of creator, team_admins, team, org_admins, org. Team and org reports: who receives each run.
connector_slugsstring[]noExternal connectors the AI analysis may use (e.g. notion, linear).
statusstringnoOne of draft, active, paused. draft, active (publish), or paused.

list_dashboards​

Read

List the dashboards the current user can see in a workspace, pinned first, with cursor pagination. Pass a renderable dashboard's id to get_dashboard to read its live data, including saved Tracked Time views, Timesheets, Personal Productivity, and widget boards.

ParameterTypeRequiredDescription
org_idstringyesWorkspace (billing) ID from get_current_user.
firstnumbernoPage size (default 50, maximum 100).
afterstringnoend_cursor from the previous page.

get_dashboard​

Read

Read a dashboard's live data with the viewer's permissions: saved Tracked Time totals and groups, Timesheets member/day stats, Personal Productivity metrics, or configured widget data. Use these numbers instead of reconstructing the dashboard with time-entry queries. Dates default to the saved range or this week. Data is returned as Markdown and JSON; respect truncation markers and use provided totals.

ParameterTypeRequiredDescription
idstringyesDashboard ID from list_dashboards.
start_datestringnoInclusive start date (YYYY-MM-DD) to override the saved range.
end_datestringnoInclusive end date (YYYY-MM-DD) to override the saved range.
start_timestringnoExact start timestamp (ISO8601), paired with end_time; do not combine with date overrides.
end_timestringnoExact exclusive end timestamp (ISO8601), paired with start_time.

create_dashboard​

Write

Create a preset dashboard for the user: productivity (personal), or timesheets, velocity, and client_health for a team. Team kinds need team_id and team-review rights. List existing dashboards first with list_dashboards to avoid duplicates.

ParameterTypeRequiredDescription
org_idstringyesWorkspace (billing) ID from get_current_user.
kindstringyesOne of productivity, timesheets, velocity, client_health. Preset kind: productivity is personal; the rest are team dashboards.
team_idstringnoTeam ID — required for timesheets, velocity, and client_health.
namestringnoOverride the preset's default name.
iconstringno
colorstringno
pinnedbooleannoPin to the dashboard nav (default true).

update_dashboard​

Write

Update a dashboard the user owns or admins through its team: rename, restyle, repin, retarget a team, or replace the widget board layout. Get the dashboard id from list_dashboards or get_dashboard.

ParameterTypeRequiredDescription
idstringyesDashboard ID from list_dashboards or get_dashboard.
namestringno
descriptionstringno
iconstringno
colorstringno
pinnedbooleannoPin to the dashboard nav.
columnsnumbernoWidget board column count.
visibilitystringnoOne of private, team, workspace.
team_idstringno
widgets_configurationrecord[]noWidget board layout; at most 40 widgets.

delete_dashboard​

Write

Delete a dashboard the user owns or admins through its team. Reports linked to it keep running on their last filters. Get the dashboard id from list_dashboards or get_dashboard.

ParameterTypeRequiredDescription
idstringyesDashboard ID from list_dashboards or get_dashboard.

list_routine_runs​

Read

List routine run summaries for the current user, ordered by most recent first. Use get_routine_run for full briefs. Filter by routine ID or status (pending, running, ready, failed).

ParameterTypeRequiredDescription
routine_idstringnoFilter runs to a specific routine ID.
statusstringnoOne of pending, running, ready, failed. Filter by run status.
limitnumbernoMax runs to return. Default 25.
cursorstringnoPagination cursor from previous response.

get_routine_run​

Read

Get a single routine run by ID, including the parent routine metadata and all briefs the run produced with their markdown bodies.

ParameterTypeRequiredDescription
idstringyesRoutine run ID.

run_routine​

Write

Run a routine now. routine_id is the routine's ID from list_routine_runs (routine.id). Team and org routines run for every member, so confirm with the user first. Returns the run with page_path, the page that shows it; check it with get_routine_run. The run sends its output to the routine's configured email or Slack destinations.

ParameterTypeRequiredDescription
routine_idstringyesThe routine ID from list_routine_runs (routine.id).

list_routines​

Read

List the user's routines (scheduled or event-triggered agent jobs) with their prompt, trigger, delivery, connectors, and status (draft, active, paused). Also returns routine_connectors: the connectors a routine can use and whether the user has connected each one (connect in Settings > Agent > Connectors).

ParameterTypeRequiredDescription
statusstringnoOne of draft, active, paused. Filter by status.

list_routine_templates​

Read

List routine templates to start a routine from with create_routine: time tracking and firm-type templates for agencies, law and accounting firms (law_firm), and professional services. Each has a suggested schedule, delivery, and connectors. Also returns routine_connectors with the user's connection status.

ParameterTypeRequiredDescription
categorystringnoOne of personal, time_tracking, agency, law_firm, professional_services. Only templates in this category.

create_routine​

Write

Create a routine: an agent job that runs on a schedule or event and delivers a brief in-app, by email, or in Slack. Start from a template_key (list_routine_templates) or write a prompt. Other fields override the template. New routines are drafts and never run on schedule until published with update_routine status active (or create with status active when the user asked to turn it on).

ParameterTypeRequiredDescription
template_keystringnoTemplate key from list_routine_templates.
namestringnoRoutine name.
promptstringnoWhat the agent should do each run.
cadencestringnoOne of daily, weekday, weekly, monthly, interval. Schedule cadence. Set schedule_enabled false for on-demand.
schedule_enabledbooleannoWhether the routine runs on its schedule.
schedule_time_of_daystringnoLocal time HH:MM, e.g. 09:00.
schedule_day_of_weeknumbernoWeekly cadence, single day: 0 = Sunday, 1 = Monday ... 6 = Saturday. Prefer schedule_days_of_week.
schedule_days_of_weeknumber[]noWeekly cadence: the weekdays it fires on, 0 = Sunday ... 6 = Saturday. [1,2,3,4,5] = Monday to Friday.
schedule_day_of_monthnumbernoMonthly cadence: 1-28, or -1 for the last day.
schedule_interval_minutesnumbernoInterval cadence, in minutes.
trigger_eventstringnoOne of transcript_finalized, meeting_ended, day_ended, time_spent, untagged_time, budget_threshold. Fire on a Rize event instead of the schedule (personal routines only). null clears it. time_spent, untagged_time and budget_threshold need trigger_config.
trigger_configrecordnoEvent settings. time_spent: {dimension: app|website|label|client|project|task, target: app name, domain, or label/client/project/task id, threshold_minutes, period: today|this_week}. untagged_time: {threshold_minutes, period}. budget_threshold: {threshold_percent, client_id?}.
delivery_channelsstring[]noAny of in_app, email, slack. Where each run's brief is delivered.
connector_slugsstring[]noExternal connectors the routine may use (e.g. notion, linear, clickup, github). See routine_connectors from list_routines.
web_accessbooleannoPublic web search on runs. null follows the workspace setting (off unless a workspace admin turns on Web access in Settings > Agent); false turns it off. Never on when the workspace has it off.
statusstringnoOne of draft, active. draft (default) or active.
runs_forstringnoOne of personal, team, org. Who it runs for (default personal). Team and org need admin rights.
team_idstringnoTeam ID when runs_for is team.
org_idstringnoWorkspace ID when runs_for is org.

update_routine​

Write

Edit a routine's name, prompt, trigger (schedule or event), delivery channels, or connectors. Omitted fields stay unchanged. Set status active to publish a draft, paused to stop it, or draft to unpublish. Team and org routines need admin rights.

ParameterTypeRequiredDescription
routine_idstringyesRoutine ID from list_routines.
namestringnoRoutine name.
promptstringnoWhat the agent should do each run.
cadencestringnoOne of daily, weekday, weekly, monthly, interval. Schedule cadence. Set schedule_enabled false for on-demand.
schedule_enabledbooleannoWhether the routine runs on its schedule.
schedule_time_of_daystringnoLocal time HH:MM, e.g. 09:00.
schedule_day_of_weeknumbernoWeekly cadence, single day: 0 = Sunday, 1 = Monday ... 6 = Saturday. Prefer schedule_days_of_week.
schedule_days_of_weeknumber[]noWeekly cadence: the weekdays it fires on, 0 = Sunday ... 6 = Saturday. [1,2,3,4,5] = Monday to Friday.
schedule_day_of_monthnumbernoMonthly cadence: 1-28, or -1 for the last day.
schedule_interval_minutesnumbernoInterval cadence, in minutes.
trigger_eventstringnoOne of transcript_finalized, meeting_ended, day_ended, time_spent, untagged_time, budget_threshold. Fire on a Rize event instead of the schedule (personal routines only). null clears it. time_spent, untagged_time and budget_threshold need trigger_config.
trigger_configrecordnoEvent settings. time_spent: {dimension: app|website|label|client|project|task, target: app name, domain, or label/client/project/task id, threshold_minutes, period: today|this_week}. untagged_time: {threshold_minutes, period}. budget_threshold: {threshold_percent, client_id?}.
delivery_channelsstring[]noAny of in_app, email, slack. Where each run's brief is delivered.
connector_slugsstring[]noExternal connectors the routine may use (e.g. notion, linear, clickup, github). See routine_connectors from list_routines.
web_accessbooleannoPublic web search on runs. null follows the workspace setting (off unless a workspace admin turns on Web access in Settings > Agent); false turns it off. Never on when the workspace has it off.
statusstringnoOne of draft, active, paused. draft, active (publish), or paused.