ONES

Execute a workflow on a specific issue. This is the only valid way to update the issue state. Before executing a workflow, please make sure to get the executable workflows using tool `get_issue_executable_workflows`. The issue's state can only be one of the valid target states defined in the executable workflows. If the issue is currently in a start state, it will transite to the end state after executing the workflow.

Find similar issues based on the content of a specific issue. The issueID is required. The targetProjectID and targetIssueTypeID are all required. If not clear, ask the user to provide them. This tool will find issues that are similar in title plus description to the specified issue within the target project and issue type.

Get wiki pages that the current user is watching or favoriting.

Get a history of what happened to specific issues, including field values changes, comments, status changes, etc. The sinceDateTime parameter determines the lower bound of activities to retrieve. This tool call is rather expensive, so only call it when necessary, and pass no more than 5 issueIDs at a time. The result is a map of issueID to activities.

Get code-related information for a specific issue, including linked commits, branches, and pull requests.

Get details of a specific issue. estimateVariance is the hours variance of the estimate. It's calculated as estimated - (actual + remaining). Any attachments are returned as attachment metadata in the result. fieldValues may include object-link / dependency fields (field_type_uuid multi_task_link or single_task_link, e.g. predecessor/successor/related links); values are related issue IDs. Discover those field UUIDs via get_issue_fields when needed. historyDisplayIds is the read-only list of retained historical semantic issue IDs, in server order with duplicates and the current ID excluded; no history is []. The list uses retained identifier/move records and does not reconstruct missing events.

Get the executable workflows for a specific issue. Executable workflow is the only valid way to update the issue state. Each workflow contains the start state, end state, and the workflow ID.

Get issue fields (uuid, name, required, types). Branches: - issueTypeID + projectID: fields from that issue type's create/detail forms in the project, plus team object-link/dependency fields (multi_task_link/single_task_link). - issueTypeID only: fields from that issue type's team-level create/detail forms, plus team object-link/dependency fields. - projectID only: union of fields across all issue types in the project, deduplicated (browse permission), plus object-link fields. required is true if any type's create form requires it; can_be_created is always false in this aggregate view. - neither: team's full issue field library (requires project administration permission); no options; can_be_created=false. Team library required flags are not project/type form required rules. Object-link fields (field_type_uuid multi_task_link or single_task_link, e.g. predecessor/successor/related links) are the way to discover work-item association relation names and UUIDs; use them with get_issue_details fieldValues or query_issues_by_onesql. Do not confuse them with hierarchy fields (field014/field114/field037/field053). Cursor-paginated over fields: page size 50; check pageInfo.hasNextPage / pageInfo.endCursor / pageInfo.total. Per-field options remain capped at 30 when included (use search_for_issue_field_options for more). Prefer value_type_desc over deprecated type: text|multi_line_text; float/duration_hour=*100000; integer/duration_second=raw; single/multi_reference_object=ID(s); date="YYYY-MM-DD"; datetime="YYYY-MM-DD HH:mm:ss" for create/update. Hyperlink (field_type_uuid is hyperlink): Prefer constructing the final wiki URL directly; Do not ask whether to use the page title or link text. Use the runtime team_uuid. {web_base_url}/wiki/#/team/{team_uuid}/page/{page_id} or /wiki/#/team/{team_uuid}/page/{page_id}. Work hours: field018 is deprecated for estimated work hours—use field060. On create, first try the simple field060 payload format; if needed, retry create_new_issue with the detailed field060 payload format (see create_new_issue). For existing issues, estimated work hours still follow the dedicated manhour/workhour tool flow after get_manhour_mode. field019/020 manhour tools only.

Get human-readable keys for a batch of issues. The response keeps the same input order and returns issueID/displayID pairs, where displayID is the issue's human-readable key.

Get all issue statuses. If projectID is provided, only issue statuses associated with that project are returned. Returns a list of issue statuses, each containing: - id: The ID of the issue status. - name: The name of the issue status.

Get issue-type hierarchy configs. Each hierarchy: name; parent_field (parent→child IDs); children_field (child→parent ID); relation_type (one_to_many|many_to_many); issue_type_hierarchy_tree with issue_type_id, issue_type_name, children[]. To link parent/child, set parent_field on child or children_field on parent (one side enough; system keeps both sides consistent).

Get issue types by optional name keyword. With projectID: returns project-associated issue types in project order (id, name, type, builtIn, issueTypeScopeID); requires project browse permission. Without projectID: returns the team issue type library sorted by name; requires project administration permission. If input is omitted or empty, return matching issue types in that scope. Cursor-paginated: page size 50; check pageInfo.hasNextPage / pageInfo.endCursor / pageInfo.total.

Get a list of comments for a specific issue. Each comment includes plain text in `text` and `markdown` converted from embeds when present; prefer `markdown` when reading the comment body. Comments may include `attachments` when embeds/files are present. When `attachments` is present, treat each item as content to extract—not metadata only. Use fileName/fileType for context; prefer an item's `summary` when available as its extracted text; for images use `base64` (data URI) when available. Do not rely only on attachment links in markdown/content when those payloads are present.

Get a list of pages in a specific wiki space(page group). Use this tool to retrieve the list of pages in a specific space by its ID. The result includes ID, title, parentID of each page and other metdata. The result contains pages hierechy information. Call this tool ONLY when you don't know how to search for a page by title OR you need the hierechy information. If you know the title keyword of the page you want to find, use the search_pages_by_title tool with spaceUUIDs parameter to limit the search scope to this space.

Get a list of manhour records in simple mode for a specific issue.

Get a list of manhour records in summary mode for a specific issue.

get manhour mode (simple or summary) for current team. Always call this tool first before any other manhour related tools to determine the manhour mode. When manhour mode is 'simple', use simple mode tools. When manhour mode is 'summary', use summary mode tools.

Get help on the ONESQL grammar and usage rules for querying issues. This tool MUST be called before using the query_issues_by_onesql tool.

Get the text content(converted to text) of a specific attachment in a wiki page.

Get the changes made to a specific wiki page since a given date-time. Returns the base version of all changes and changes(operations) grouped by user and ordered by time after the specified date-time (sinceDateTime). The returned changes include the base version of all the changes (in "data" field), the operations performed grouped by user and ordered by time (in "ops" field").

Get details of a specific page in a wiki space. Includes title, full content (markdown), comments, attachments, and metadata. When `attachments` is present, treat each item as content to extract—not metadata only. Use fileName/fileType for context; prefer an item's `summary` when available as its extracted text; for images use `base64` (data URI) when available. Do not rely only on attachment links in markdown/content when those payloads are present.

Get project details by project ID.

Get project members grouped by role. Without roleID: overview of all roles, each capped at the first 50 member IDs; memberTotal is the full count per role; use roleID+cursor for the rest. With roleID: cursor-paginated members of that role (page size 50; pageInfo.hasNextPage / pageInfo.endCursor / pageInfo.total). The same user may appear under multiple roles. Requires member component view permission.

Get project roles. With projectID: returns roles configured in the project (requires member component view permission). Without projectID: returns team-level project role definitions (requires project administration permission). Cursor-paginated: page size 50; check pageInfo.hasNextPage / pageInfo.endCursor / pageInfo.total.

Get details of a specific wiki space(page group). Use this tool to retrieve the details of a specific space by its ID. It's returns homePageID, which is the ID of the home page of the space. When you need to create a new page in root of the space, you can use this ID as the parentPageID in the create_page tool.

Get details of a specific sprint. The response includes the sprint built-in fields and custom fields.

Get details of a specific test case in a test case library.

Get all fields configured in a test case library.

Get user notices since a specific datetime in ISO 8601 format.

Post a comment on a specific issue. The comment can be used to notify users by mentioning them. To mention a user, include '@username' or '@user_id' in the comment text. Add spaces around the mention to ensure proper parsing. The full username or user ID (which can be retrieved using search_for_users tool) must be used for the mention to work correctly.

Query issues using ONESQL. ONESQL has its own grammar and usage rules. Before using this tool, get_onesql_grammar_help tool MUST be used to get the grammar and usage rules of ONESQL. This tool can be used instead of get_issue_details to query one or many issues with complex conditions and get fields' values flexibly.

Read attachment content by resourceID. Use this tool when another tool returns attachment's resourceID and you need attachment content for model reading. Important: - This tool returns attachment content to the model context; it does not save files to local working directories. - Supported outputs are text (utf-8 decode) and images. - Binary formats like video/audio/archive are not returned by this tool. In that case, tell the user this is a content-read tool, not a local file download tool, and suggest a different download path if local files are expected.

Search for epics by name keyword and optional project ID. If input is not provided, returns all epics accessible to the user. Max return 50 items, if cannot find the epic you want, please provide more specific keywords.

Search for options of a specific issue field based on a keyword. Normally the options are retrieved from get_issue_fields tool. But when there are too many options, only 30 options are returned by that tool. This tool can be used to search for more options. Or some options are not returned, you can also use this tool to search for the missing options. Max return 50 items, if cannot find the option you want, please try refining the keyword and try again. Some built-in fields that can be searched with this tool: - field029: product field. The options are the list of products can be associated with the issue. - field030: product module field. The options are the list of product modules that can be associated with the issue. Some other module-like fields use different tools: - field054: project module field. Use search_for_modules tool instead. (fields that are not listed above may or may not be searchable with this tool, please try and see if it works when necessary.)

Search for modules by name keyword and optional project ID. If input is not provided, returns all modules accessible to the user. Max return 50 items, if cannot find the module you want, please provide more specific keywords.

Search for modules in a test case library by name keyword. If input is not provided, returns modules accessible in the target library. Max return 200 items, if cannot find the module you want, please provide more specific keywords.

Search for projects by partial name and additional conditions. If input is not provided, returns all projects with applied conditions and accessible to the user. Max return 50 items, if cannot find the project you want, please provide more specific keywords.

Search for references in ONES wiki using the original question and five expanded questions. When user ask questions that needs wiki knowledge/corporate knowledge, use this tool to search for references. Five expanded questions must be generated from the original question and provided to improve search results. Three of them are rephrased versions of the original question from different angles, One is a more general question (a step back version) of the original question, The last one is a possible answer to the original question. Use spaceIDs / pageIDs to limit the search scope to specific wiki spaces or pages when necessary. spaceIDs and pageIDs are mutually exclusive, only one of them can be provided at a time.

Search for releases (publish versions) by name keyword and optional project ID. If input is not provided, returns all releases accessible to the user. Max return 50 items, if cannot find the release you want, please provide more specific keywords.

Search for wiki spaces by a keyword. This tool returns a list of wiki spaces that match the given keyword, including their ID, name, and other metadata. Returns 50 spaces at most. Refine your keyword if you need more specific results. If input is not provided, it returns the first 50 spaces in the team.