ContextLayer
ContextLayer: граф знаний, сессии, задачи и документы для команд с ИИ
От сообщества: Добавлен пользователем или импортирован; проверьте владельца перед подключениемРаботаетНужен API-ключГлобальныйБесплатноТолько чтение
Что умеет
- Создавать проекты, спринты и задачи
- Назначать задачи участникам и отслеживать зависимости
- Искать и получать документы по контексту
- Отслеживать активность и статус проекта в реальном времени
- Управлять командами и должностями в организации
Какие данные видит
Доступ выдаётся по API-ключу и ограничен организацией и проектами, к которым привязан пользователь.
Нужен ли аккаунт
Нужен API-ключ из настроек сервиса
ContextLayer — инфраструктура контекста для команд, работающих с ассистентами: граф знаний, управление проектами и задачами, сессии и документы. Ассистент может создавать проекты и спринты, назначать задачи, отслеживать зависимости и искать документы по контексту.
Как подключить
Как подключить ContextLayer к Claude
Адрес сервера, который нужно вставить в настройках Claude:
https://api.dotnova.io/mcp
- Откройте Claude (сайт claude.ai или приложение для компьютера).
- Перейдите в Настройки → Коннекторы (в английском интерфейсе: Settings → Connectors).
- Нажмите Добавить пользовательский коннектор (Add custom connector).
- Вставьте адрес сервера, скопированный ниже, в поле URL удалённого MCP-сервера и нажмите Добавить.
- В чате нажмите + → Коннекторы и включите новый коннектор.
Пользовательские коннекторы доступны на тарифах Free, Pro, Max, Team и Enterprise. На Free можно добавить только один. В Team и Enterprise коннектор сначала добавляет владелец организации в Настройки организации → Коннекторы.
Авторизация
После нажатия Добавить откроется окно входа в сервис. Войдите под своим аккаунтом и подтвердите доступ. Пароль Claude не видит.
Как подключить ContextLayer к ChatGPT
Адрес сервера, который нужно вставить в настройках ChatGPT:
https://api.dotnova.io/mcp
- Откройте ChatGPT в браузере (chatgpt.com). Нужен тариф Plus, Pro, Business, Enterprise или Edu.
- Один раз включите режим разработчика: Настройки → Приложения → Дополнительные настройки → Режим разработчика (Settings → Apps → Advanced settings → Developer mode).
- Откройте Настройки → Коннекторы и нажмите Создать (Create).
- Заполните форму: Название (любое), Описание (одна фраза, что делает сервис), URL MCP-сервера (скопируйте ниже).
- В поле Аутентификация выберите OAuth, если сервис требует вход, иначе Нет. Нажмите Создать.
- В новом чате откройте + → Приложения/Коннекторы и включите коннектор.
Название раздела у OpenAI менялось (Connectors → Apps/Plugins). Если пункт называется иначе, ищите «режим разработчика» в настройках. В рабочих пространствах Business/Enterprise администратор должен разрешить пользовательские коннекторы.
Авторизация
При первом использовании ChatGPT откроет окно входа в сервис. Войдите и подтвердите доступ.
Как подключить ContextLayer к Cursor
Адрес сервера, который нужно вставить в настройках Cursor:
https://api.dotnova.io/mcp
Быстрый способ: нажмите кнопку Открыть в Cursor ниже, Cursor сам предложит добавить сервер.
Вручную:
- В Cursor откройте Settings → Cursor Settings → MCP и нажмите Add new global MCP server.
- Вставьте JSON, скопированный ниже, в открывшийся файл
~/.cursor/mcp.json(для одного проекта:.cursor/mcp.jsonв корне проекта). - Сохраните файл. Сервер появится в списке MCP; при необходимости пройдите авторизацию по кнопке рядом с ним.
Авторизация
Если сервис требует вход, рядом с сервером в списке MCP появится кнопка авторизации.
Список инструментов сервера (62)
Технические названия из tools/list. Нужны только разработчикам.
| accept_org_invite | Accept a pending organization invite by organization name. |
| add_task_dependency | Add a dependency between tasks. Use this to express that a task is blocked by another task (must complete first) or relates to it. This prevents context contamination: when Task B depends on Task A, agents working on Task B will be warned if Task A is still in progress. |
| add_team_member | Add a member to a team. Find the team by name or ID, and the user by name or email. |
| assign_task | Assign a user to a task by name or email. The user is added to the task's assignees and linked in the knowledge graph. |
| create_position | Create a new job position/title in your organization (admin only). |
| create_project | Create a new project in your organization. You will be added as the project owner. |
| create_sprint | Create a new sprint for a project. Automatically migrates non-completed tasks (todo/in_progress) from the previous sprint. On first use, creates Sprint 0 (Backlog) for existing tasks and Sprint 1 as the active sprint. |
| create_task | Create a new task in a project. Tasks track work items and can be assigned to team members and linked to sessions. |
| create_team | Create a new team in your organization. You will be added as team lead. |
| delete_document | Delete a document from the knowledge base. This permanently removes the document. |
| delete_position | Delete a position permanently (admin only). Find by title or ID. |
| delete_project | Delete a project permanently (admin only). Find by name or ID. |
| delete_task | Delete a task permanently. The task and all its assignee links are removed. |
| delete_team | Delete a team permanently (admin only). Find by name or ID. |
| delete_user | Delete (soft-delete) a user from your organization (admin/owner only). |
| end_session | End a work session. This triggers context block generation — your activities are summarized and stored in the knowledge graph for future reference. |
| find_documents | Search documents in the organization's knowledge base. Filter by category, tags, project, or free-text query. Returns document metadata — use get_document to read the full content. |
| get_context | Get intelligent context from the knowledge graph. Pass a query describing what you need — the system will automatically route to the right data sources (tasks, sessions, documents, teams, blocks) based on intent analysis + graph traversal + semantic similarity. Optionally narrow scope with project name or session ID. |
| get_context_routing | Get the current context routing mode for the organization. Returns 'keyword_llm', 'keyword_only', or 'llm_only'. |
| get_document | Get the full content of a document by its ID. Use find_documents first to search, then get_document to read the content. For binary documents (PDFs, images), content is returned base64-encoded with encoding='base64' and the mime_type field. |
| get_live_activity | See what's happening right now: active sessions, recent events, traces, and tasks being worked on. |
| get_my_stats | Get your personal activity statistics: total sessions, events, active days, projects worked on, and recent activity summary. |
| get_org_stats | Get comprehensive organization statistics: counts, engagement metrics, task velocity. |
| get_project_status | Get the current status of a project including recent activity. You can use either the project name or ID. |
| get_session_details | Get full details of a session including all events/activities logged during it. Shows who did what, when, and the session block summary if available. Admins can view any session in the org; members can only view their own. |
| get_team_details | Get detailed information about a team: members (with roles), linked projects, and recent activity. Use team name or ID. |
| invite_user | Invite a new member to your organization (admin/owner only). |
| link_task_session | Link a work session to a task. This records that the session was used to work on the task, connecting them in the knowledge graph. |
| link_team_project | Link a project to a team. Find both by name or ID. |
| list_colleagues | List colleagues in your organization. Shows name, email, role, and optionally filters by team. Admins see all users; members see users in their own teams. |
| list_notifications | List your notifications — task assignments, status changes, sprint updates, invites, and more. Returns unread notifications by default. |
| list_pending_invites | List pending organization invites for the current user. |
| list_positions | List all job positions/titles in your organization with member counts. |
| list_project_members | List all members involved in a project. Includes users who have sessions in the project and users from teams linked to the project. |
| list_project_teams | List teams linked to a project. |
| list_projects | List projects in your organization. No parameters needed — just call it to see all your projects. |
| list_sessions | List your recent work sessions. Optionally filter by project name or ID. Shows task type, status, and event count. |
| list_sprints | List sprints for a project. Shows sprint number, name, goal, status (current/completed), and task count. |
| list_tasks | List tasks. Filter by project, status (todo/in_progress/done/cancelled), sprint number, or show only your tasks. Shows title, status, priority, assignees, sprint, and linked sessions. |
| list_teams | List all teams in your organization. Shows team name, member count, project count, and description. |
| log_activity | Log an activity or event. If no session_id is provided, a session is automatically created or reused for the project. Just pass project_name and content — no need to start a session first. |
| log_trace | Log a granular agent action trace. High-frequency, no LLM processing. |
| log_traces_batch | Log multiple traces in a single batch. |
| mark_notification_read | Mark a notification as read, or mark all notifications as read at once. |
| move_task_to_sprint | Move a task to a different sprint. Specify the target sprint by number or ID. |
| remove_task_dependency | Remove a dependency between tasks. This unblocks the task from the other task. |
| remove_team_member | Remove a member from a team. Find the team by name or ID, and the user by name or email. |
| remove_user_from_org | Remove a user from the current organization without deleting their account. Admin/owner only. |
| search_entities | Search entities in the knowledge graph by name. Finds people, companies, documents, concepts, tools, etc. that have been mentioned in your activities. |
| set_context_routing | Set the context routing mode for the organization. Requires admin role. Options: 'keyword_llm' (balanced), 'keyword_only' (fastest), 'llm_only' (most accurate). |
| setup_organization | Set up your organization in one step (admin only). Use this for onboarding. |
| start_session | Start a work session. If a project is provided (project_name or project_id), the session is bound to it; if neither is provided, an unbound (projectless) session is created — useful when the work isn't tied to a specific project. Reuses an existing active session for the same scope (per-project, or one global projectless session per user). Task type: coding, review, planning, debugging, research, meeting, other. |
| unassign_task | Remove a user from a task's assignees. |
| unlink_team_project | Unlink a project from a team. Find both by name or ID. |
| update_org | Update your organization's details (owner only). Set name, description, business model, objectives, custom context for AI, or links. |
| update_position | Update a position's title, description, or permissions (admin only). Find by title or ID. |
| update_project | Update a project's name, description, or links. Find by name or ID. |
| update_task | Update a task's status, title, description, priority, or team assignment. Use this to move tasks through the workflow (todo -> in_progress -> done) or reassign to a team. |
| update_team | Update a team's name, description, or objectives. Find by name or ID. |
| update_user | Update an existing user's name, role, positions, or skills (admin/owner only). |
| upload_document | Upload a document to the organization's knowledge base. Documents can be templates, policies, contracts, procedures, guides, or references. They are indexed in the knowledge graph and automatically surfaced as context when relevant to a task. Supports both text and binary files: for binary files (PDFs, images, etc.), set is_base64=true and provide the content as a base64-encoded string, along with the appropriate mime_type (e.g. 'application/pdf', 'image/png'). Optionally set file_name for the original filename. Embeddings are generated automatically for search. |
| whoami | Get your profile, organization, projects, recent sessions, and top entities. This is the best starting point — call this first to understand what's available. No parameters needed. |