sitedrive_create_work_packages
ChatGPTCreate up to 100 ordinary packages with at most 1000 scheduled locations in total. Discover work_area_id with location summaries/search; use null for unplaced work. Each package needs a name and at least one location with offset-qualified planned_start and planned_end. Unplaced work cannot be mixed with placed locations. New packages always follow the project default location breakdown, including children of parents using another breakdown. Every placed location must belong to the project default breakdown, or the discovered space-fallback breakdown when no default is configured. Placed areas must map to schedulable spaces; parent areas without their own mapping and empty areas reject the whole batch. Use null for unplaced work; the NIL UUID is rejected. Optional parent_id must belong to this project; adding children to a leaf parent replaces its scheduled tasks with child work. Optional order_number sets package order among siblings (finite decimals or null); omission leaves it unset. UI sorting can rewrite sibling orders. No schedule view is required. Returns package and task-definition IDs in request order; save these for later updates. external_id is a customer mapping field, not a uniqueness or retry key. Never automatically retry after a lost response: creation may have committed. Read package summaries before deciding whether to retry. Uses existing project write permissions. Changes are broadcast to open schedules.
sitedrive_get_skill
ChatGPTLoad one focused skill by ID from sitedrive_list_skills or a tool description. Returns the server-owned SKILL.md and its content revision; the identical document is available at its resource URI. Read move-scheduled-work for package/phase/location/void moves and proportional resize. Instructions guide tool use and do not expand user authorization.
sitedrive_get_space_model
ChatGPTRead a project's space model: its buildings, sections, floors, zones and spaces as a tree, parents before their children, each with its path and whether work is placed on it. Use the ids with sitedrive_save_space_model. Show the user the result as an indented outline of names with their kind in words ("Kh · Kitchen"), never ids; collapse long runs of similar siblings ("… 40 more stretches"), and end with the url. At most 5000 spaces per answer; next_cursor continues the list.
sitedrive_get_work_package_history
ChatGPTInspect retained history for one known work package or phase, including its descendants. First use work-package summaries for an overview and to select the package ID; if no phase or package is specified, ask the user to choose one and explain that a project-wide activity report is unsupported. Do not enumerate packages or phases to simulate one. Optional virtual_space_id and virtual_space_namespace_id narrow the selection to a schedule work area and its breakdown, using the same selection as the work-package drawer. Returns newest events first, recorded planned-start, planned-end and progress changes with package, task and area names, and nullable author IDs. Values are the saved new values, not before/after pairs. Responses fit a 48 KB JSON budget, with at most 20 changes per event and 5 task/area names per change. details_truncated, omitted_changes, omitted_area_names, omitted_task_names and text_truncated explicitly mark excerpts. Narrow the package/area or lower limit to recover detail; omitted detail is not on later pages. Do not treat omitted details or empty-change events as proof of no edits. Some events have no detailed changes; this is not a complete audit of all fields or user activity. Never guess an unknown author or infer intent or delay causes. For a date-range question, use the project timezone from project search, pass from (inclusive) and before (exclusive). Date and user_id filters run in the database before pagination; use sitedrive_search_users to resolve a named author, or its ids lookup to resolve returned author IDs. For yourself use sitedrive_whoami. Keep the same from and user_id on later pages. Use next_cursor unchanged as before. A partial page does not establish a total or prove a person made no changes. Author IDs may be unavailable for system or ambiguously attributed events; do not infer them from package assignees.
sitedrive_get_work_sections
ChatGPTRead one work section library as a tree: every section with code, name, description, parent_code, order_number and depth, in display order (depth-first, siblings by order_number). Codes are unique within a library and are how the write tools address sections. read_only is true for locked libraries and Sitedrive system templates. Any account member can read. Write tools require system_config access as well as admin rights on the library's account; a write to another account gets access_denied. Locked libraries and Sitedrive system templates are read-only for everyone (namespace_read_only); unlock a library in the Sitedrive UI first.
sitedrive_list_location_breakdowns
ChatGPTList a project's location breakdowns: ways of grouping the space model's leaf spaces into work areas, such as by floor or into takt areas. Automatic ones follow the space model's levels and cannot be edited. Give breakdown_id to read one breakdown's areas in tree order, each with its spaces and its work_area_id, and the spaces in no area. Show the user the areas as an indented outline of names with their spaces, never ids; collapse long runs of similar areas ("… 40 more takt areas"), say how many spaces are in no area, and end with the url.
sitedrive_list_schedule_views
ChatGPTList a project's schedule views (the tabs people open the schedule in) in menu order, with each view's kind, who it is shared with and a link that opens it. Use the ids with sitedrive_save_schedule_views and sitedrive_remove_schedule_views. Views can also hold filters, grouping and columns set in the app; those are not shown and never change through these tools.
sitedrive_list_skills
ChatGPTDiscover focused Sitedrive skills before an unfamiliar multi-tool task. Returns metadata only; use sitedrive_get_skill with an item ID to load instructions. Filter by domain (for example scheduling) or a literal substring in its ID or description. Follow next_cursor until null when browsing the complete catalog.
sitedrive_list_work_package_dependencies
ChatGPTList the work package dependencies that decide which work must finish before other work can start. Each item reads as: work_package_name waits for prerequisite_name, by rule, plus lag_working_hours. Filter by successor work_package_id, prerequisite_id or both (they combine with AND); omit both for the whole project. Find package IDs with sitedrive_list_work_package_summaries. space_mappings are set only for SPACE_TO_SPACE and use work_area_id values from location summaries. One answer lists at most 200 pairs in total, in list order: from the first dependency whose pairs do not fit, space_mappings is null, with space_mapping_count still given. To read it, filter by its work_package_id and prerequisite_id. A dependency with more than 200 pairs cannot be read whole here, and setting it replaces all its pairs, so edit its existing pairs through the Sitedrive REST API. next_cursor continues the list; keep filters unchanged.
sitedrive_list_work_package_location_summaries
ChatGPTFind current schedule work by floor/location, person, team, dates or package. First resolve the project with sitedrive_search_projects, a physical floor with sitedrive_search_locations, and a named crew with sitedrive_search_teams when needed. For “what is unfinished on Level 3?”, use the physical space ID as space_id and completion=incomplete. Returns one complete package/location aggregate per item, not nested package totals. Filters select complete package–location summaries; their progress includes all contributing tasks. Includes package and location names. Use package summaries for additional package hierarchy context. Use assigned_user_id=me for my work. A floor match may include a work area extending beyond that floor. Returns a bounded answer; has_more means narrow the question. Do not interpret a bounded result as an exact total. Never enumerate the whole schedule through repeated calls. Use REST summary resources for full extraction. Use response timezone and explicit offsets for calendar weeks. Starts next week: planned_start_from/planned_start_before. Use completion for task-state selection, overdue_as_of for unfinished work past planned end, and sort_by=progress with limit=10 for least-complete locations. These are location results, not whole-package overdue totals. Do not infer readiness, delay causes, forecasts or historical progress. Working duration is planned calendar span, not labor effort.
sitedrive_list_work_package_summaries
ChatGPTBrowse current schedule packages/phases and their whole-package planned/actual dates, weighted progress and planned working duration. Use parent_id=null for a top-level overview or a package ID for a focused phase summary. Returns package IDs, parent IDs and names in structural order. Every package includes has_location_rows regardless of fields: false means skip its location-detail lookup; true includes eligible work in leaf descendants and unplaced work, but additional location filters may still return no rows. This is neither a location count nor an indication of child packages. Request a parent by its ID only when additional hierarchy context is needed. No location summaries. Use sitedrive_list_work_package_location_summaries for work on a floor, personal/team assignments or date filters. Use returned IDs to narrow follow-up requests. Never average displayed child progress to obtain parent totals. Working duration is planned calendar span, not labor effort. Returns a bounded answer; has_more means narrow search or package scope. Use REST summary resources for full extraction.
sitedrive_list_work_section_namespaces
ChatGPTList the work section libraries (WORK_SECTION namespaces) of one account, the classifications work packages are tagged with: id, name, description, is_default, locked, archived and read_only. Get account_id from sitedrive_whoami. Sitedrive system templates are not listed. Any account member can read. Write tools require system_config access as well as admin rights on the library's account; a write to another account gets access_denied. Locked libraries and Sitedrive system templates are read-only for everyone (namespace_read_only); unlock a library in the Sitedrive UI first.
sitedrive_move_work_package_to
ChatGPTMove one scheduled package, phase or location occurrence to planned_start using the project working calendar, preserving its working duration. Use an explicit timezone offset and a valid working-calendar start. For discovery, scope, calendar interpretation and read-back verification, load sitedrive_get_skill with id move-scheduled-work. Find package IDs using package summaries. Omit work_area_id to affect a whole package or phase, including descendants; use work_area_id from location summaries for one occurrence, or null for void work. Requires project admin update rights; schedule planners are not supported for this release. Unknown, deleted, foreign-schedule, or unmatched occurrence IDs make the entire selection a no-op; read the result and verify IDs if nothing changed. After a lost response, read current dates before deciding whether to retry. Sequence and parallelization are not supported.
sitedrive_move_work_packages
ChatGPTMove and proportionally resize scheduled work using the project working calendar. For discovery, scope, calendar interpretation and read-back verification, load sitedrive_get_skill with id move-scheduled-work. Find package IDs using package summaries. Omit work_area_id to move a whole package or phase, including descendants; use work_area_id from location summaries for one occurrence, or null for void work. move_by shifts selected packages by working_seconds, preserving spacing and duration. Repeat a work_package_id to select distinct work_area_id occurrences in one move_by request, including null for void work. Duplicate selections and overlapping affected work are rejected. move_to positions one target at planned_start. resize_start or resize_end proportionally resizes one target while preserving the other edge. Use explicit timezone offsets and working-calendar dates. Requires project admin update rights; schedule planners are not supported for this release. Unknown, deleted, foreign-schedule, or unmatched occurrence IDs make the entire selection a no-op; read the result and verify IDs if nothing moved. Never retry after a lost response: relative moves may have committed. Sequence and parallelization are not supported.
sitedrive_move_work_packages_by
ChatGPTShift selected scheduled work by signed working_seconds using the project working calendar, preserving durations and relative spacing. Negative seconds move earlier. Accepts up to 100 non-overlapping selections; repeat a package ID for distinct work_area_id occurrences, including null for void work. Duplicate selections and overlapping affected work are rejected. Never automatically retry a relative move after a lost response: it may have committed. For discovery, scope, calendar interpretation and read-back verification, load sitedrive_get_skill with id move-scheduled-work. Find package IDs using package summaries. Omit work_area_id to affect a whole package or phase, including descendants; use work_area_id from location summaries for one occurrence, or null for void work. Requires project admin update rights; schedule planners are not supported for this release. Unknown, deleted, foreign-schedule, or unmatched occurrence IDs make the entire selection a no-op; read the result and verify IDs if nothing changed. After a lost response, read current dates before deciding whether to retry. Sequence and parallelization are not supported.
sitedrive_remove_schedule_views
ChatGPTRemove 1–50 schedule views by id from sitedrive_list_schedule_views. Only the view goes; the schedule's work is untouched. A view with removable false is refused, and at least one view must stay shared with everyone. Ids that are no longer in the project are skipped, so repeating a call is safe; if a removal fails partway, the error lists the views already removed. Creating views and changing their name or kind needs project admin or planner rights; changing order or sharing of an existing view, and removing views made by others, needs project admin rights. Returns the removed and skipped ids and the schedule url.
sitedrive_remove_work_package_dependencies
ChatGPTRemove 1–100 dependencies between work packages, each named by work_package_id (the package that waits) and prerequisite_id. A dependency that does not exist is skipped. Unknown, deleted or other-project package IDs reject the whole batch. Only project admins can remove dependencies to packages without scheduled work. Returns the removed dependencies and the schedule url. Requires project admin, or schedule planner responsible for every listed package. The whole batch succeeds or fails together; changes reach open schedules live. Repeat-safe: after a lost response, run it again or read the dependencies back.
sitedrive_remove_work_packages
ChatGPTRemove up to 100 whole project packages and ALL their scheduled locations. Destructive operation using existing soft-deletion/history behavior. Every selected ID must be an existing project package without children; unknown, deleted, foreign IDs and parents reject the entire batch without partial deletion. Selecting a parent together with its children is also rejected. Locked packages cannot be removed. Returns removed IDs. A repeated successful removal is rejected because the IDs are already deleted. Uses existing project write permissions. Changes are broadcast to open schedules.
sitedrive_resize_work_package_end
ChatGPTProportionally resize one scheduled package, phase or location occurrence to planned_end while keeping its planned start fixed. Uses the project working calendar. Use an explicit timezone offset and a valid working-calendar finish. For discovery, scope, calendar interpretation and read-back verification, load sitedrive_get_skill with id move-scheduled-work. Find package IDs using package summaries. Omit work_area_id to affect a whole package or phase, including descendants; use work_area_id from location summaries for one occurrence, or null for void work. Requires project admin update rights; schedule planners are not supported for this release. Unknown, deleted, foreign-schedule, or unmatched occurrence IDs make the entire selection a no-op; read the result and verify IDs if nothing changed. After a lost response, read current dates before deciding whether to retry. Sequence and parallelization are not supported.
sitedrive_resize_work_package_start
ChatGPTProportionally resize one scheduled package, phase or location occurrence to planned_start while keeping its planned finish fixed. Uses the project working calendar. Use an explicit timezone offset and a valid working-calendar start. For discovery, scope, calendar interpretation and read-back verification, load sitedrive_get_skill with id move-scheduled-work. Find package IDs using package summaries. Omit work_area_id to affect a whole package or phase, including descendants; use work_area_id from location summaries for one occurrence, or null for void work. Requires project admin update rights; schedule planners are not supported for this release. Unknown, deleted, foreign-schedule, or unmatched occurrence IDs make the entire selection a no-op; read the result and verify IDs if nothing changed. After a lost response, read current dates before deciding whether to retry. Sequence and parallelization are not supported.
sitedrive_save_location_breakdown
ChatGPTCreate or replace one custom location breakdown: its name and all of its areas, each a group of the space model's leaf spaces. Send the whole breakdown every time: areas left out are removed. List each area by its path of names ("Track" › "Takt 1"); parents need not be listed, and siblings keep the listed order. To keep a stored area (and its work_area_id) while renaming or moving it, give its id. An area with child areas holds no spaces; every other area holds at least one. A parent space stands for the leaf spaces under it, and a space sits in at most one area. Spaces may stay in no area. Work stays on its spaces: changing areas only regroups it. Automatic breakdowns cannot be changed. Repeat-safe. Requires project admin. Show the user the areas as an indented outline of names with their spaces, never ids; collapse long runs of similar areas ("… 40 more takt areas"), say how many spaces are in no area, and end with the url.
sitedrive_save_schedule_views
ChatGPTCreate 1–50 schedule views, or change their name, kind, menu order and sharing. Omit id to create a view (name and kind required); give an id from sitedrive_list_schedule_views to change one, where omitted fields stay as they are. Names are unique within the project, so repeating a create is refused instead of making a duplicate. A view with renamable false keeps its name and kind. At least one view must stay shared with everyone. Find people to share with using sitedrive_search_users. Filters, grouping and columns set in the app are kept. Every item is checked before anything is written; if a later write still fails, the error lists what was already saved. Creating views and changing their name or kind needs project admin or planner rights; changing order or sharing of an existing view, and removing views made by others, needs project admin rights. Returns each view with a link that opens it.
sitedrive_save_space_model
ChatGPTCreate or change 1–5000 spaces in a project's space model, the location tree its location breakdowns are built from. Name a new space by key and its parent by parent_key, so a whole tree fits in one call; change a stored space by id. Give every new space a property_type, your best guess at what kind of place it is. Only the fields you send change: a stored space keeps every field you leave out, and null clears it. Never removes a space. Adding a space under one that has work moves that work down into the new space, and moving a space takes its work along: say so and ask first when has_work is true. Repeat-safe. Requires project admin. Returns each saved space with its path and kind, and a url to the space model in Sitedrive. Show the user the result as an indented outline of names with their kind in words ("Kh · Kitchen"), never ids; collapse long runs of similar siblings ("… 40 more stretches"), and end with the url. Levels typed SECTION, FLOOR or ZONE each get a location breakdown, and one breakdown lists every space: say which ones the tree gives.
sitedrive_search_locations
ChatGPTResolve a named floor, room or work area within a known project. By default search covers both physical spaces and persisted work areas, regardless of space_type: older floors may be typed ZONE. Supply a non-empty name search. Names and ancestor paths disambiguate matches; do not choose the first duplicate. For a physical floor or room, use the space result as space_id; use a work_area result as work_area_id only when the user means that schedule breakdown. Same-named spaces and work areas can contain different work, so ask which scope when unclear. A selected work area may extend beyond the named floor. has_more means narrow the search.
sitedrive_search_projects
ChatGPTFirst step for finding a schedule: search project, site or account names directly, without calling sitedrive_whoami. A site can contain several schedules with unrelated names; show candidates and ask which one when ambiguous. Search is required and literal, case-insensitive. Narrow by account_id only when already known. Results include names, IDs, timezone, match source and a project link. Only readable active projects appear. has_more means refine the name or account. A known project ID from a schedule URL can go directly to a project-scoped tool.
sitedrive_search_teams
ChatGPTFind a crew by name on a known project site, or resolve returned team_ids in one bounded IDs lookup. Supply a name or IDs. Use returned IDs in project-scoped team filters. has_more means narrow the name or IDs.
sitedrive_search_users
ChatGPTResolve a person by literal, case-insensitive name, email or phone on a known project site, or resolve returned author IDs with a bounded ids lookup. Supply search or ids; supplying both intersects the filters. For yourself use sitedrive_whoami. Phone matching preserves formatting; use a shared digit substring. Searches and limits in the backend within the caller’s existing user visibility. Returns user IDs for author filters in work-package history and assigned_user_id in location summaries. Only listed site/project members appear; account-wide access alone does not imply site membership. A result is not proof they edited or are assigned to the selected work. Ask which person when names are ambiguous; never choose an arbitrary match. has_more means refine the search or split the requested IDs. IDs missing from the result remain unresolved; do not guess their names. An empty result does not prove the person does not exist: visibility may restrict the directory to the caller.
sitedrive_set_work_package_dependencies
ChatGPTCreate or update 1–100 dependencies saying which work must finish before other work can start. work_package_id waits for prerequisite_id. Pick the rule from how the work follows on site; for guidance load sitedrive_get_skill with id work-package-dependencies. Setting an existing dependency replaces its rule, lag and mappings; other dependencies stay unchanged. Find package IDs with sitedrive_list_work_package_summaries and work_area_id values with sitedrive_list_work_package_location_summaries. Dates do not move by themselves: the planner sees the dependencies in the schedule. Returns the saved dependencies and the schedule url to view them. Requires project admin, or schedule planner responsible for every listed package. The whole batch succeeds or fails together; changes reach open schedules live. Repeat-safe: after a lost response, run it again or read the dependencies back.
sitedrive_set_work_package_progress
ChatGPTSet current reported progress for the TASK work in a package or phase. First find the project and package, then use location summaries if the user named a specific work area. Omit location_id to update every descendant leaf and location; pass one exact location_id for that work area, or null for unplaced work. This is a real schedule mutation and may affect actual dates; ask the user before calling when intent is ambiguous. Every selected TASK instance receives the same percentage, while displayed package progress remains weighted. At 100%, legacy prerequisite tasks are completed too. Never retry automatically after a timeout; read the current summary first.
sitedrive_submit_mcp_feedback
ChatGPTWhen Sitedrive MCP cannot meet a user goal after trying the relevant tools, explain the gap and offer to submit feedback. Call this tool only if the user explicitly asks you to submit it or accepts your offer. Do not submit for missing permissions, a transient error, or a legitimate empty result. The short feedback text is stored in Sitedrive logs and analytics for product review. Never include secrets or a full chat transcript.
sitedrive_update_work_package_metadata
ChatGPTUpdate name, description, abbreviation, hex_color, order_number, work_section_id and external_id for up to 100 project packages atomically. Supply each id and at least one field. Omitted fields stay unchanged; null clears nullable fields. name cannot be null or blank. Does not change dates, progress, hierarchy or location assignments: use the dedicated date/progress tools for those. Uses existing project write permissions. Changes are broadcast to open schedules.
sitedrive_whoami
ChatGPTReturns the authenticated Sitedrive user and their accounts, including a link to each account in the app and whether MCP tools are enabled on it. Use for identity, authentication or access questions, e.g. when a tool says MCP is not enabled. To find a schedule, call sitedrive_search_projects directly; its search does not need an account ID.
sitedrive_create_work_packages
Claudesitedrive_get_skill
Claudesitedrive_get_work_sections
Claudesitedrive_list_skills
Claudesitedrive_list_work_package_location_summaries
Claudesitedrive_list_work_package_summaries
Claudesitedrive_list_work_section_namespaces
Claudesitedrive_move_work_package_to
Claudesitedrive_move_work_packages_by
Claudesitedrive_remove_work_packages
Claudesitedrive_resize_work_package_end
Claudesitedrive_resize_work_package_start
Claudesitedrive_search_locations
Claudesitedrive_search_projects
Claudesitedrive_search_teams
Claudesitedrive_set_work_package_progress
Claudesitedrive_submit_mcp_feedback
Claudesitedrive_update_work_package_metadata
Claudesitedrive_whoami
Claude