anamnese

Create a task. For recurring tasks, provide freq and recurrence pattern fields.

One-off task fields: deadline, scheduled_date, has_scheduled_time, has_deadline_time Recurring task fields (require freq): scheduled_time, deadline_offset_days, deadline_time, days_of_week, days_of_month, week_position, months, interval Do NOT mix one-off and recurring fields (e.g. scheduled_date + scheduled_time is invalid).

Recurrence examples:

• Every weekday at 9am: freq="weekly", days_of_week=[1,2,3,4,5], scheduled_time="09:00"

• Every 3 days: freq="daily", interval=3

• Monthly on the 15th: freq="monthly", days_of_month=[15]

• First Monday of month: freq="monthly", week_position=1, days_of_week=[1]

Frequency rules:

• daily: No modifier fields needed

• weekly: days_of_week required [0=Sun..6=Sat]

• monthly: Either days_of_month OR (week_position + single day_of_week)

• yearly: months [1-12] + days_of_month required

Use scheduled_date for when to work on it, deadline for when it must be done. Time support: Both deadline and scheduled_date support full ISO datetime. Set has_scheduled_time=true when scheduling at a specific time.

Permanently delete a goal.

This is a hard delete - the goal cannot be recovered. Use update_goal with status='achieved' if you want to mark a goal as achieved instead.

Permanently delete a memory. This cannot be undone.

Permanently delete a note.

Use when:

• Knowledge is no longer relevant

• Note was created in error

• Information is outdated and should be removed

Delete a task. For recurring tasks, use apply_to to control scope.

• "this" (default): For recurring tasks, skips this occurrence (prevents re-showing as virtual). For one-off tasks, permanently deletes.

• "all_future": Deletes the recurrence template. Stops generating new instances. Already-completed instances are preserved.

Retrieve the full content of a specific note by ID.

Use after search_notes to load the detailed content of a note you need. Returns the complete note including full content (up to 10,000 chars).

PRIMARY TOOL - Call this at the START of every conversation to load comprehensive user context.

Returns:

• current_datetime: Current date and time in the user's timezone (ISO 8601 with offset)

• All active facts about the user (preferences, personal info, relationships)

• tasks_overdue: Tasks with scheduled_date OR deadline in the past

• tasks_today: Tasks scheduled OR due today (time >= now), plus unscheduled tasks (no date set)

• tasks_tomorrow: Tasks scheduled OR due tomorrow (includes projected recurring tasks)

• Active goals

• Recent moments from the last 5 days

• Latest 15 user-facing notes (id + description). Use get_note to retrieve full content.

• ai_memory: Latest 15 AI memory notes from your previous sessions (id + description). Use get_note to retrieve full content.

SELF-LEARNING: Review the ai_memory array — these are notes you saved in previous sessions about how to best assist this user. Load relevant ones with get_note. Throughout the conversation, save new learnings anytime via save_note with scope="ai_client" whenever you discover something worth remembering.

• tasks_recently_completed: Tasks completed or skipped in the last 7 days

Each task includes:

• category_reason: 'scheduled' | 'deadline' | 'both' - explains why it's in that array

• has_scheduled_time: true if task has a specific scheduled time, false if all-day

• has_deadline_time: true if deadline has a specific time, false if all-day

Task placement uses scheduled_date when present, otherwise deadline. Each task appears in exactly one category.

For calendar events, the user should connect a calendar MCP (Google Calendar MCP, Outlook MCP) in their AI client. Query those MCPs alongside Anamnese for a complete daily view.

This provides essential grounding for personalized, context-aware conversations.

Create a new goal for the user.

Goals are aspirational objectives that represent longer-term ambitions. They can be personal or professional.

Examples:

• Personal: "Learn Spanish to conversational fluency", "Run a marathon"

• Professional: "Get promoted to senior engineer", "Launch my own product"

Save a memory about the user.

type "fact" = Stable truths (preferences, personal info, relationships). Stays true for months/years. Examples: "User is a software engineer", "User prefers TypeScript", "User's dog is named Max"

type "moment" = Time-bound events, decisions, experiences. Include occurred_at when timing matters. Examples: "User decided to prioritize X", "User had a meeting about Q1 OKRs"

Save a new note with learned knowledge or procedures.

Notes store knowledge you learn during conversations that might be useful later:

• How to do something in this codebase/project

• Procedures, configurations, or technical details

• Solutions to problems encountered

• Project-specific knowledge

Notes have two parts:

• description: Short summary for searching (max 500 chars)

• content: Detailed knowledge (max 10,000 chars)

Use notes for LEARNED KNOWLEDGE. Use facts for TRUTHS ABOUT THE USER.

Examples:

• description: "How to deploy this Next.js project to Vercel" content: "1. Run 'vercel' command... 2. Configure environment variables..."

• description: "Database migration process for this project" content: "Migrations are in supabase/migrations/. To apply: npx supabase db push..."

SELF-LEARNING (scope="ai_client"): Your persistent memory across conversations. Save a note whenever you learn something worth remembering — don't wait, save as you go. Examples:

• User preferences: "User prefers concise answers, not long explanations"

• Corrections: "User clarified: 'deploy' means push to staging, not production"

• Interaction patterns: "User likes to review plans before I execute"

• What works or doesn't: "Suggesting refactors unprompted frustrates this user"

The more you learn and remember, the better you become at helping this user.

Search user goals with optional filters.

Use to find goals by keyword, type (personal/professional), status (active/achieved), or tags. Without filters, returns goals by recency.

Search user memories by keyword, type, tags, or date range.

With query: Case-insensitive keyword search on content. Without query: Returns memories by recency. Use type filter to search only facts or only moments.

Search notes by keyword or list recent notes. Returns summaries (id + description) only. Use get_note to retrieve the full content of a specific note.

With query: Case-insensitive keyword search on description and content. Without query: Returns most recently updated notes.

Search tasks with optional filters. Returns both real tasks and projected recurring task instances.

When a date range is provided (scheduled_date, scheduled_date_from/to), virtual recurring instances are automatically included. Virtual tasks have is_virtual=true and IDs starting with "virtual:". They can be passed directly to update_task or delete_task.

Filter conflicts: scheduled_date and scheduled_date_from/to are mutually exclusive. unscheduled=true cannot be combined with date filters.

Use to find tasks by keyword, date, priority, tags, or completion status. Use without filters to get recent tasks.

Update an existing goal's fields.

When status is set to 'achieved', achieved_at is automatically set. When status is set to 'active', achieved_at is automatically cleared.

Use to:

• Update goal content

• Change goal type

• Mark goal as achieved or reactivate

• Change tags

Update an existing memory's content or tags.

Works for both facts and moments. Use to correct info, update tags, or expand content.

Update an existing note's description or content.

Use this to:

• Update or expand knowledge in an existing note

• Fix or improve the description for better searching

• Add new details learned about a topic

Update a task. Accepts real task IDs or virtual recurring task IDs (from search_tasks).

For recurring tasks, use apply_to to control scope:

• "this" (default): Update only this specific occurrence. Materializes virtual tasks automatically.

• "all_future": Update the recurrence template. Changes affect all future occurrences.

Instance fields (apply_to="this" ONLY): deadline, scheduled_date, has_scheduled_time, has_deadline_time, percent_complete, is_skipped Template fields (apply_to="all_future" ONLY): freq, interval, days_of_week, days_of_month, week_position, months, scheduled_time, deadline_offset_days, deadline_time, is_active Both modes: title, content, priority, duration_minutes, tags Note: is_skipped=true and percent_complete=100 are mutually exclusive.

Common operations:

• Complete a task: percent_complete=100

• Skip a recurring occurrence: is_skipped=true

• Reschedule: scheduled_date="new-date"

• Pause a recurrence: apply_to="all_future", is_active=false

• Change recurrence pattern: apply_to="all_future", freq=..., days_of_week=...