All servers / CRM / ViviScape MCP server
ViviScape MCP server
MCP server for ViviScape Work: CRM, projects, tasks, time, notes, artifacts and insights, authenticated as the signed-in ViviScape Work user or a scoped agent c
Tested by ConnectATDScore 86 out of 100. Tested 2026-11-19 on package version 2.26.0. Package: viviscape-mcp · source
Score breakdown
| Runs | 18 / 20 | |
|---|---|---|
| Tool descriptions | 16 / 20 | |
| Package safety | 20 / 20 | |
| Maintenance | 20 / 20 | |
| Adoption | 12 / 20 |
What we saw
- Starts and answers the MCP handshake: yes (2.7 seconds including download)
- Tools offered: 155
- Last published: 2026-11-16
- Weekly downloads (npm): 1,178
- License: MIT · Direct dependencies: 3
Want this server set up for your team?
ConnectATD builds AI automation and wires up MCP servers like this one. Tell us what you are trying to do.
Talk to ConnectATDTools and how they are described
Agents choose tools from these descriptions. Under each name you can see how many of its parameters are explained.
auth_statusno parametersShow the ViviScape session this server is using (user, account, role, plan)
auth_loginno parametersSign in to ViviScape Work through the browser and store the session for this server
auth_logoutno parametersClear the stored ViviScape session
prospect_add15 of 15 parameters describedAdd a new prospect/lead to SalesFront, the ViviScape CRM. Give an email, a phone or a mobile (at least one). Pass tags to label it on creation (RECON, an industry, a campaign) -- they are applied right after the prospect is created, as prospect_tags add would.
prospect_update3 of 11 parameters describedUpdate an existing SalesFront prospect. Only the fields you pass change; the rest are kept (the current record is read first and merged). Use prospect_tags for tags and prospect_note_add for notes.
prospect_get1 of 1 parameter describedGet a prospect by ID
prospect_query4 of 4 parameters describedSearch/query prospects with filters
prospect_list_by_accountno parametersGet all prospects for the current account
prospect_list_by_rep1 of 1 parameter describedGet all prospects assigned to a specific sales rep
prospect_remove1 of 1 parameter describedRemove a prospect from the CRM
prospect_tags5 of 5 parameters describedList, add, or remove color-coded tags on a SalesFront prospect -- how leads are grouped for recon and outreach (RECON, WARM INTRO, an industry, a campaign). Same rules as task_tags: labels are stored upper-case and add is idempotent per label (re-adding one recolors it and returns the existing tag). add returns the persisted tag including its tag_id; remove takes that tag_id, not the label. A prospect on another account comes back as an empty tag (tag_id all zeros). To go from a label to the prospects carrying it, use prospects_by_tag; prospect_query rows also carry tag_items, and its search text matches tag names.
prospects_by_tag3 of 3 parameters describedFind the account prospects carrying given tags -- the reverse of prospect_tags, and how a recon list is read back (every lead marked RECON, everything tagged both HEALTHCARE and WARM INTRO). Labels match trimmed and case insensitively. match "any" (default) returns a prospect carrying at least one label; "all" narrows to prospects carrying every one. Rows are light (prospect_id, name, org, email, phone, stage, rep, created) and carry the whole tag set. Pass no tags to list the account tag vocabulary instead: every distinct label with its color and prospect count.
prospect_note_add3 of 3 parameters describedAdd a note to a prospect
prospect_note_update2 of 2 parameters describedUpdate a prospect note
prospect_notes_list1 of 1 parameter describedGet all notes for a prospect
prospect_note_remove1 of 1 parameter describedRemove a prospect note
prospect_followup_add6 of 6 parameters describedLog a follow-up (a touch: an email sent, a call, a meeting) on a SalesFront prospect. Defaults to now when no date is given.
prospect_followup_update1 of 6 parameters describedUpdate a followup record
prospect_followups_list1 of 1 parameter describedGet all followup records for a prospect
company_listno parametersList all companies
company_add1 of 8 parameters describedAdd a new company
company_update1 of 8 parameters describedUpdate an existing company
client_get1 of 1 parameter describedGet a client by ID
client_get_by_email1 of 1 parameter describedLook up a client by email address
client_add5 of 6 parameters describedAdd a new client to a company
clients_by_company1 of 1 parameter describedGet all clients for a company
project_list4 of 4 parameters describedGet all projects for current user. Returns a page of trimmed rows; see fields/limit/offset.
project_list_active4 of 4 parameters describedGet active projects for current user. Returns a page of trimmed rows; see fields/limit/offset.
project_get1 of 1 parameter describedGet a project by ID
project_staff1 of 1 parameter describedGet staff members assigned to a project
project_tasks12 of 12 parameters describedGet all tasks for a project. Returns a page of trimmed rows; see fields/limit/offset.
tasks_open14 of 14 parameters describedGet open tasks across projects. company_id and only_mine filter server-side; the rest filter in-process. Returns a page of trimmed rows with computed due_date / due_in_days / overdue -- read those, not the API deadline string, which is unreliable.
task_add17 of 17 parameters describedAdd a task to a project. The API requires a full Group_Task payload; this tool builds it from simplified params.
task_update9 of 10 parameters describedUpdate a task. Fetches existing task first and merges changes. This is where an agent records its plan -- see the plan argument.
task_get1 of 1 parameter describedGet a single task by ID
task_tags5 of 5 parameters describedList, add, or remove tags on a task -- the categorisation primitive of the task board, and the way to mark work an agent created or touched. Labels are stored upper-case, and add is idempotent per label: re-adding one returns the existing tag rather than a duplicate. add returns the persisted tag including its tag_id; remove takes that tag_id, not the label, and answers true. This tool is per task -- list rows from tasks_open / project_tasks carry tags: null, so read a task tag set here or with task_get. To go the other way, from a label to the tasks carrying it, use tasks_by_tag.
task_references5 of 5 parameters describedList, add, or remove cross references between tasks -- the way related work is linked so one task points at another (a duplicate, a blocker, the ticket a task came from). Links are SYMMETRIC: adding A -> B makes the pair visible from both tasks, so link once and never mirror it by hand. list rows describe the OTHER task -- related_task_id plus its title, status and project -- so no second task_get is needed to report them. add is idempotent per pair and returns the persisted link including its reference_id; remove takes that reference_id, not a task id, and answers true. Both tasks must belong to the signed-in account. Use task_search to find the id to link when only a keyword is known.
task_search3 of 3 parameters describedKeyword search over the account tasks and tickets -- matches the title, the description, and, for a numeric query, the task id itself. Returns light rows (task_id, title, status, project, company), enough to pick an id for task_get or task_references without pulling a full task feed.
tasks_by_tag4 of 4 parameters describedFind the account tasks carrying given tags -- the reverse of task_tags, and the way a tag-driven queue is read (every task an agent marked, everything labelled RELEASE-BLOCKER, and so on). Labels match trimmed and case insensitively, so DEVOPS, devops and " DevOps " are the same tag. match "any" (default) returns a task carrying at least one label; match "all" narrows to tasks carrying every one. Rows are light (task_id, title, status, project, company) and carry the task whole tag set, so the other labels come back with the hit -- no follow-up task_tags call. Pass no tags to list instead the account tag vocabulary: every distinct label with the number of tasks using it, which is how to discover what is worth filtering on.
tasks_pending13 of 13 parameters describedGet pending tasks for a user. Returns a page of trimmed rows; see fields/limit/offset.
tasks_by_milestone13 of 13 parameters describedGet tasks for a milestone. Returns a page of trimmed rows; see fields/limit/offset.
tasks_by_group_user14 of 14 parameters describedGet tasks by project group and user. Returns a page of trimmed rows; see fields/limit/offset.
tasks_by_company12 of 12 parameters describedGet tasks for a company. Returns a page of trimmed rows; see fields/limit/offset.
projects_by_company6 of 6 parameters describedGet user projects filtered by company. Returns a page of trimmed rows; see fields/limit/offset.
projects_active_by_user5 of 5 parameters describedGet active projects for a specific user. Returns a page of trimmed rows; see fields/limit/offset.
notes_account6 of 6 parameters describedList notes across the account, newest-first as the API returns them. Bodies are omitted by default (hundreds of rows); pass fields:["note_id","title","note"] to read content.
note_revisions1 of 1 parameter describedGet the edit history of a note. Use before rewriting a shared note so human edits are not silently clobbered.
note_companies4 of 4 parameters describedLink a note to a company, list its links, or remove one. This is how a generated document gets filed against a client. NOTE: add returns an echo with zeroed ids, so call list afterward to get the assignment_id you need for remove.
note_users5 of 5 parameters describedShare a note with a user, list who it is shared with, change a permission, or unshare. NOTE: add returns an echo with zeroed ids; call list to get the assignment_id.
note_tags4 of 4 parameters describedList, add, or remove tags on a note -- the only way to categorise notes for later retrieval. NOTE: remove answers false even when it succeeds; re-list to confirm.
media_center15 of 15 parameters describedThe account Media Center (document library, also used as AI knowledge). Actions: list, get, add, update, remove. Only add a document here when a person asks for it in the Media Center: it is visible account-wide and feeds AI answers. Do NOT use it to hand files to another agent or share working files in a room; use room_files for that. add takes exactly one file source: content_base64 + filename, content_text (+ filename), an https source_url, or file_path (local installs only). update changes only the fields you pass. remove deletes the file for everyone in the account.
note_attachments5 of 5 parameters describedList, upload, or remove note attachments, or flag one as knowledge-base material. upload takes a path to a local file.
notebook6 of 6 parameters describedNotebooks group notes into a per-client or per-project documentation space. Actions: list, add, update, remove, notes (list a book contents), assign / unassign notes. WARNING: remove is broken server-side (400 even for an empty book) -- delete notebooks in the ViviScape Work UI.
notebook_users5 of 5 parameters describedList, add, update, or remove the users a notebook is shared with
project_files12 of 12 parameters describedList, upload, update, delete, or resolve the download URL of files attached to a project. upload takes a path to a local file; delete is by file_id from list.
task_files5 of 5 parameters describedList files attached to a task. Use this to read context humans already attached before generating anything new.
services7 of 7 parameters describedResolve service ids to names. Task rows carry a service_id and frequently a null service_name, so this is how a caller turns 356 into "Feature Request". Actions: list (all), active, get (one by id), by_company.
user_lookup4 of 4 parameters describedResolve a user id, email, or team. Assignees arrive as bare ids (210, 230, 264), so use this rather than guessing who they are. Actions: by_id, by_email, exists, team (companies a user belongs to), groups (their projects).
tasks_bulk_add4 of 4 parameters describedCreate many tasks in one request. Prefer this over looping task_add: the API builds each task server-side, so it is one round trip and one failure surface. All tasks land in the same project with the same starting status.
enumsno parametersList the status, priority, and group-type values the platform actually accepts. Read this before setting a status or priority; tasks complete as "completed" while projects complete as "complete", and priority is low/moderate/important/urgent/critical.
project_add7 of 7 parameters describedCreate a project for a company. NOTE: the API has no project-delete route, so a project created here can only be archived or renamed afterward.
project_update2 of 2 parameters describedUpdate a project. Reads the current row and merges your changes onto it, so only pass what changes. Field names are the raw GEN_Group ones (strGroupName, strGroupDescription, bitPublic, target_date, ...).
project_set_status3 of 3 parameters describedSet a project status without touching its budget or approved time. Project status values are new, inprogress, complete -- note "complete", not "completed".
project_user_add4 of 4 parameters describedAdd a staff member to a project, optionally as lead or client contact
project_user_remove2 of 2 parameters describedRemove a staff member from a project
milestone_list1 of 1 parameter describedList milestones for a project
milestones_activeno parametersList active milestones across the account
milestones_by_company1 of 1 parameter describedList milestones for a company
milestones_by_user1 of 1 parameter describedList active milestones for a user (defaults to the signed-in user)
milestone_get1 of 1 parameter describedGet a milestone by ID
milestone_add6 of 6 parameters describedCreate a milestone in a project. Use this to give tasks a phase to hang off; tasks reference it via milestone_id.
milestone_update8 of 8 parameters describedUpdate a milestone. Reads the current record and merges your changes onto it.
milestone_clone3 of 3 parameters describedCopy an existing milestone into a project. Useful for repeatable delivery templates.
milestone_remove1 of 1 parameter describedDelete a milestone
task_assignee_add5 of 5 parameters describedAssign a user to a task
task_assignee_remove1 of 1 parameter describedUnassign a user from a task. Takes the assignee_id from the task staff_assignees list, NOT a user_id.
task_set_leader2 of 2 parameters describedMake a user the leader on a task
task_set_attention3 of 3 parameters describedSet or clear the attention flag for a user on a task
task_delete1 of 1 parameter describedDelete a task permanently. There is no undo; prefer setting status to cancelled unless the task was created in error.
task_merge2 of 2 parameters describedMerge one task into another. The source task is consumed by the target; use this to clean up duplicates.
task_complete3 of 3 parameters describedClose a task through the platform completion route. Prefer this over task_update with status "completed": it also stamps who finished the work and runs the milestone/progress rollup the task board reads.
tickets_list9 of 9 parameters describedThe account ticket queue, filtered server-side by status set, company and team. status "open" (default) covers backlog, new, research, inprogress, testing, inreview and onhold; "closed" reads the completed/cancelled route; "all" drops the status filter. Tickets are tasks, so rows are task rows -- read one with task_get and search by keyword with task_search. Requires the enterprise plan.
flow_listno parametersList the ViviScape Flows you can start right now (published and live). Each lists its inputs: the fields it reads out of the input you send (name, type, required). Start one with flow_start.
flow_start5 of 5 parameters describedStart a ViviScape Flow as yourself. Name it by flow_id (from flow_list) or by its exact name. input is what the flow receives: an object with the fields flow_list shows (or input_text for plain text). Pass external_id to make a retry safe: the same id is recorded once. Returns the request (status received/running/done/failed/ignored) and its run when one started; check progress with flow_status.
flow_status1 of 1 parameter describedThe status of a flow request you started (flow_start): the request, and its run with status, outcome, error and each step reached.
task_comments1 of 1 parameter describedGet all comments on a task. Task rows report total_comments and unread_for_me; this reads the thread itself.
task_comment_get1 of 1 parameter describedGet a single task comment by ID
task_comment_add4 of 4 parameters describedAdd a comment to a task. WARNING: on this platform account the backend fires a client-notification webhook for comments on tasks whose client_id > 0, so a comment can reach the customer. Check the task client_id first if that matters.
task_comment_update4 of 4 parameters describedUpdate the text of an existing task comment
task_comment_remove2 of 2 parameters describedRemove a task comment by ID. Pass task_id: the backend answers 500 even when the delete succeeds, and task_id lets the removal be verified instead of reported as a failure.
task_mark_read1 of 1 parameter describedMark a task read for the signed-in user, clearing unread_for_me
room_listno parametersList the conference/chat rooms on the signed-in platform account, with each room's users
room_get1 of 1 parameter describedGet a room by ID: name, greeting, PIN, active flag and users
room_add4 of 4 parameters describedCreate a room (Space) on the signed-in platform account. The backend generates the dial-in PIN and adds the signed-in user as the room admin. An agent credential needs the rooms.create permission: the Space is made in the agent's account, the person who owns the agent becomes its admin (client_id is ignored), and the agent is granted the new Space so it can read, post and share files there right away. Returns the room_id.
room_update4 of 4 parameters describedRename a room, deactivate/reactivate it, or change recording. Only room_name, active and enable_recording can change; omitted fields keep their current values. There is no room delete: deactivate instead.
room_messages1 of 1 parameter describedRead the chat messages in a room. Each row has chat_id, user_id, name, message, post_date, agent_key/agent_name when an AI agent posted it, and mentions (who it tags). To be told about NEW messages without polling, use room_subscribe.
room_message_add5 of 5 parameters describedPost a chat message to a room as the signed-in user. Tag people or agents by writing their @handle in the text (room_mentionables lists them) or by passing mentions; the result lists who was tagged.
room_message_update5 of 5 parameters describedEdit the text of an existing room chat message
room_message_remove2 of 2 parameters describedDelete a room chat message by ID. Pass room_id: the backend answers 500 even when the delete succeeds, and room_id lets the removal be verified instead of reported as a failure.
room_mentionables1 of 1 parameter describedWho can be tagged in a room: the people and active AI agents of its account, each with the @handle that tags them (an agent's name or a person's first+last name, letters and digits only, e.g. @Gee, @ArtHicks). Write @handle in a message, or pass mentions to room_message_add.
webhook_catalogno parametersThe events a webhook can send, by scope (Spaces, Tasks, Tickets) and group (the item itself, comments, assignees, tags, files, members, messages), each with its created / updated / removed events, plus delivery limits.
webhook_listno parametersThe account's outbound webhooks: address, events, Space / project filters, whether an auth header is set (name + hint, never the value), status (active / paused, with why), and delivery health.
webhook_add6 of 7 parameters describedCreate an outbound webhook: every matching event is POSTed as JSON to url (https), with X-ViviScape-Event / -Event-Id / -Delivery-Id headers and the optional auth header. Non-2xx answers are retried for about a day; 50 failures in a row pause it. Send webhook_action "test" afterwards to check the receiver.
webhook_update9 of 9 parameters describedChange a webhook. Only the fields given change; event_types replaces the whole list. auth_header_value replaces the stored secret; auth_header_name alone renames the header and keeps the value; clear_auth_header removes it. [] for room_ids / project_ids removes that filter.
webhook_action1 of 2 parameters describedpause (stop sending; events wait), resume (send again, resets the failure streak), remove (deletes it and its secret; waiting events are dropped), or test (send a webhook.test event now and return the receiver's status code, answer and time).
webhook_deliveries4 of 5 parameters describedA webhook's delivery log, newest first: event type, status (pending / sending / delivered / failed / skipped), attempts, the receiver's last status code or error. With delivery_id: that one delivery including its JSON payload; add redeliver=true to send a finished one again.
room_subscribe5 of 5 parameters describedSubscribe to a room's new chat messages so they are PUSHED to you (via Azure Service Bus) instead of polling room_messages. In a local (stdio) session started as a Claude Code channel (VIVISCAPE_CHANNEL=1), each new message arrives in the conversation as a <channel source="viviscape" ...> event; otherwise receive them with the subscription tool, action "wait". Idempotent: the same room, options and label return the same subscription. Your own messages are not delivered unless include_own. An agent credential needs read access (rooms.read and a grant) to the room. Events wait up to a day; a subscription nobody receives from for 7 days expires (pushing renews it automatically).
subscription5 of 5 parameters describedManage your event subscriptions (from room_subscribe). Actions: "list" -- your subscriptions (include_inactive for removed/expired); "wait" -- block until new events arrive (up to wait_seconds, max 25) and return them, at once if some are queued. Each event is handed out once. Loop on this when push is not available; "remove" -- delete the subscription and stop delivery; "status" -- what this local session is pushing right now.
artifact_publish21 of 21 parameters describedCreate (publish) an artifact in ViviScape Work: a prototype (HTML, with its CSS/JS/images), a report or document (Markdown), an image, a PDF, a file, or a data object (JSON, XML, CSV, EDI, email). Attach it to a room (Space), task, ticket, project, company or client with parent_type + parent_id, or leave parent_type "none" for the account library. Signed in as an agent, parent_type must be room, task, ticket, project or company (with parent_id; company needs the agent's all-projects grant). A card is posted in the room or on the task / ticket unless post_companion is false or status is "draft". Send files inline (files[]), or local_path for a file or a whole folder (local installs only). kind is worked out from the files when omitted. Use artifact_update to publish a new version of the same artifact instead of publishing a duplicate. Limits: 25 MB and 200 files per version; executables are refused.
artifact_update8 of 15 parameters describedChange an artifact. With files or local_path it publishes a NEW VERSION (the same card shows it; earlier versions are kept); without them it changes the envelope only: title, summary, metadata, correlation_id, format, schema, and -- for the author or an account admin -- visibility and status (archived / published; archive also stops its share links).
artifact_archive2 of 2 parameters describedArchive an artifact: it leaves the default lists, its share links stop working and its cards say it is archived, but every version is kept. Pass restore: true to bring an archived artifact back (published; its share links work again). Only its author (you, if you published it) or an account admin can archive or restore it. Prefer this over artifact_delete.
artifact_delete2 of 2 parameters describedDelete an artifact for good: every version and file is removed, its share links stop working, and its cards say it is no longer available. Only its author (you, if you published it) or an account admin can delete it. To hide it but keep it, use artifact_update with status "archived" instead.
artifact_get6 of 6 parameters describedRead an artifact: its envelope (kind, parent, author, version, source, format, schema, metadata, correlation, lineage) and file list. include_content returns one file (the entry file, or path) as text, or base64 for binary files. versions lists every version; view_url returns a short-lived link to the sandboxed viewer (about 30 minutes) for a person to open.
artifact_move5 of 5 parameters describedMove an artifact: attach it to another room (Space), task, ticket, project, company or client, or to the account library (parent_type "none"); an agent can move it only to a room, task, ticket, project or company. Who can see it follows the new place; its versions, share links and history come along. Only its author or an account admin can move it, and you must be able to publish where it goes. A card is posted where it lands (rooms and tasks; never on a ticket) unless post_companion is false. Cards already posted where it was stay.
artifact_share_in5 of 5 parameters describedPost an existing artifact's card into another room (Space) or task, with an optional message. The artifact does NOT move and who can see it does not change: people there who cannot already see it get "not available" (e.g. it is private, or it is attached to a room they are not in -- use artifact_move for that). Not into tickets: the customer sees every ticket comment. Drafts and archived artifacts cannot be shared.
artifact_list4 of 9 parameters describedList artifacts you can see, newest first: everything in the account, or those on one room / task / ticket / project / company / client (parent_type + parent_id). within_type + within_id lists everything UNDER a project (it and its tasks / tickets) or a company (it, its clients, projects and tasks / tickets). Filter by kind, a title/summary search, or correlation_id.
artifact_share8 of 9 parameters describedShare links for an artifact, for people outside the account. list shows every link and its views; create makes a link -- "restricted" (anyone holding it; optional password) or "public" (anyone on the internet, no password) -- that follows the latest version or pins one, with an optional expiry; the full URL is returned ONCE, so pass it on right away. revoke turns a link off immediately. Only the artifact's author or an account admin manages its links; an agent credential needs the artifacts.share permission and manages links only on artifacts it published. Inside the account, the work_link from artifact_get already opens it for anyone who can see it -- use that instead when no outside link is needed.
room_files14 of 14 parameters describedShare files between agents and people inside a Space (room). This is the place to hand another agent an image, draft, export or any working file: Space files are stored apart from the Media Center and never appear there. Actions: list, add, get, remove. add takes exactly one source: content_base64 + filename, content_text (+ filename), an https source_url, or file_path (local installs only); pass message to also post a chat note so the room sees it. get returns text files as text and other files as base64 (up to 2 MB), or writes to save_path (local installs only). list also shows Media Center documents a person linked to the Space (space_file: false); remove deletes a Space file for everyone, and only unlinks a linked Media Center document (agents cannot unlink those).
timelog_add10 of 10 parameters describedLog time against a task. Duration uses shorthand format (e.g., 1h30m, 30m, 2h). Builds full LOG_Time payload.
timelog_update10 of 10 parameters describedUpdate an existing time log entry. Builds full LOG_Time payload.
task_log_time6 of 6 parameters describedLog time against a task the short way: the quick route takes the project and service from the task itself, so only the note matters. Omit duration (or pass 0) to post a note-only entry -- it lands on the task time feed without adding hours. Use timelog_add when the entry needs an explicit project, service or user.
personal_logs5 of 5 parameters describedPersonal time logs for a user -- work booked against a service rather than any task.
personal_log_add5 of 5 parameters describedAdd a personal time log: work that belongs to a service rather than to a task. For task work use task_log_time or timelog_add.
personal_log_remove1 of 1 parameter describedDelete a personal time log by its log_id (from personal_logs).
timeclock_punch3 of 3 parameters describedPunch the time clock for the signed-in user. type "in" opens a punch, "out" closes the open one. This is the attendance clock, separate from task time logs.
timeclock_history5 of 5 parameters describedTime clock punch history for a user -- clock-in/clock-out pairs with their notes.
notes_mineno parametersGet all notes for the current user
note_get1 of 1 parameter describedGet a note by ID
note_add3 of 3 parameters describedCreate a new note
note_update2 of 3 parameters describedUpdate a note
note_remove1 of 1 parameter describedDelete a note
notes_query1 of 1 parameter describedSearch notes
account_infono parametersGet account information
account_usersno parametersGet all users in the account
account_servicesno parametersGet active services for the account
insights_hours_by_person3 of 3 parameters describedGet logged hours by person
insights_timelogs2 of 2 parameters describedGet time logs for the account within a date range
insights_person_stats3 of 3 parameters describedGet stats for a person
insights_hours_by_person_service4 of 4 parameters describedGet logged hours by person and service type
insights_alltime_hours_person_service2 of 2 parameters describedGet all-time hours per person and service
insights_project_hours_by_service3 of 3 parameters describedGet hours by project broken down by service type
insights_account_hours_by_service2 of 2 parameters describedGet account hours broken down by service type
insights_ai_summary3 of 3 parameters describedGet an AI-generated summary of time logs
insights_person_hours_week1 of 1 parameter describedGet a person's logged hours for the current week
insights_person_hours_month1 of 1 parameter describedGet a person's logged hours for the current month
insights_person_hours_lastmonth1 of 1 parameter describedGet a person's logged hours for last month
insights_project_time_total1 of 1 parameter describedGet total time logged for a project
insights_all_users_timeno parametersGet time totals for all users
memory_recall4 of 4 parameters describedRecall what you have remembered about how to work: your preferences, behaviors, facts and lessons from earlier sessions. Call this at the START of a session (and before a task where a preference may apply) and follow what comes back. pinned_by_person = a person wrote or approved it: it wins over anything else you believe. Signed in as an agent this is your own memory (needs the "Its own memory" permission, memory.self); signed in as a person, pass agent_id.
memory_remember6 of 6 parameters describedRemember a preference, behavior, fact or lesson about how to work, so a later session starts knowing it. One sentence a person can read and correct ("Art wants task comments as HTML, never markdown"). Remembering the same key again updates that memory instead of adding a duplicate, so reuse keys. Remember what was said or shown to you, not guesses; put the observation in evidence. Your own memories are capped at 0.85 confidence and show on your Memory tab, where a person can correct or pin them. Not for documents or procedures longer than a sentence: write an SOP with kb_sop.
memory_forget2 of 2 parameters describedForget one memory (by memory_id from memory_recall) that is wrong or no longer applies. Signed in as an agent you can forget only your own, and not one a person pinned: ask them, or remember a correction instead.
kb_search6 of 6 parameters describedSearch the account Second Brain -- the semantic index over tickets (their description, plan, notes and solution), notes, products, distilled resolutions and SOPs. This is the memory to consult BEFORE planning work: it answers "has this been solved here before, and how". Rows carry source_type + source_key, so a hit on a ticket names the task id to open with task_get, and a hit on an SOP names the sop_id to read with kb_sop. score is 1 - cosine distance, so higher is closer. IMPORTANT: an empty result is not proof the knowledge is absent -- the route answers with no hits when Second Brain is switched off for the account or its AI token budget is inside the reserve. Check kb_status when a search of an established account comes back empty.
kb_statusno parametersReport whether Second Brain is switched on for this account and how current its index is. Read this when kb_search comes back empty on an account that plainly has history: search answers with no hits (not an error) when the brain is disabled or the AI token budget is inside its reserve, and the freshness figures show what is still queued for ingest.
kb_sop9 of 9 parameters describedList, read, or write SOPs -- the durable, deliberately authored half of the Second Brain, and the way an agent contributes knowledge meant to outlive one ticket (a runbook, a recurring fix, a house convention). Ticket and note knowledge is ingested automatically; an SOP is the only knowledge an agent authors directly. save creates when sop_id is omitted and updates in place when it is given, and it returns the sop_id either way. A saved SOP is QUEUED for indexing rather than indexed on the spot, so kb_search will not surface it until the next ingest tick. Prefer updating the existing SOP over filing a near-duplicate: search first with kb_search source_types ["sop"]. New SOPs land as status "draft" so a person can approve them.
kb_reindex2 of 2 parameters describedQueue one source document for re-indexing into the Second Brain. The portal enqueues on every write already, so reach for this only to repair a document that never made it in -- a ticket whose plan or solution kb_search cannot find, say. source_key is the task id for a ticket, the note/product/SOP GUID otherwise.
Tested in a sandbox with an empty environment: no API keys, a throwaway folder, nothing from a real account. A passing test shows the server runs and responds. It is not a security audit.