NEXO Brain

Get an agent's full profile

Register a new agent

Update agent fields

List all registered agents

Remove an agent from registry

Create an immediate backup of the NEXO database

List available backups with dates and sizes

Restore database from a backup (DESTRUCTIVE)

RAG query over cognitive memory (STM+LTM). Triggers rehearsal on retrieved results.

Cognitive memory system metrics: STM/LTM counts, strengths, retrieval stats

Inspect a specific memory by ID (debug). Does NOT trigger rehearsal.

Performance metrics: retrieval relevance %, repeat error rate, multilingual recommendation

Detect conflicts between a new instruction and established LTM memories. force=True to skip discussion.

Resolve a cognitive dissonance: paradigm_shift, exception, or override.

Detect user sentiment and get tone guidance. Also logs for tracking.

View or adjust trust score (0-100). Without args: view. With event: adjust.

Search entities by name, value, or type

Create a new entity (person, service, URL)

Update an entity's fields

Delete an entity

List all entities grouped by type

Log a code/config change with full context: what, why, trigger, affects, risks

Search past code changes — answers 'what did we change in X?' or 'why did we touch Y?'

Link a change log entry to its git commit hash

Log a non-trivial decision with reasoning, alternatives, and evidence

Record what actually happened after a past decision

Search past decisions — answers 'why did we do X?'

Show decisions and learnings that are due for review

Write end-of-session diary with decisions, discards, and context for next session

Read recent session diaries for context continuity

Search across ALL NEXO memory — changes, decisions, learnings, followups, diary, entities, .md files, code files.

Show current NEXO dimension scores (episodic memory, autonomy, proactivity, self-improvement, AGI)

Show past evolution cycles, proposals, and their outcomes

Manually trigger an evolution analysis outside weekly schedule

Approve a pending Evolution proposal

Reject a pending Evolution proposal with reason

Check learnings relevant to files/area BEFORE editing code. Call this before any code change.

Get guard system statistics: repetition rate, trends, top problem areas

Log a learning repetition (new learning matches existing one)

Get a specific preference value

Set a preference (creates or updates)

List all preferences grouped by category

Delete a preference

Register new session, clean stale ones, return active sessions + alerts.

Call this ONCE at the start of every conversation. Returns the session ID (SID) — store it for use in all other nexo_ tools.

Update session task, check inbox and pending questions.

Call this at the START of every user interaction (before doing work). Args: sid: Your session ID from nexo_startup. task: Brief description of current work (5-10 words).

Cleanly close a session. Removes it from active sessions immediately.

Call this when ending a conversation to avoid ghost sessions. Args: sid: Session ID to close.

List active sessions. Filter by keyword if provided.

Track files being edited. Detects conflicts with other sessions.

MUST call before editing any file outside ~/claude/. Args: sid: Your session ID. paths: List of absolute file paths to track.

Stop tracking files. If no paths given, releases all.

Args: sid: Your session ID. paths: File paths to release. Omit to release all.

Show all tracked files across all active sessions with conflict detection.

Send a fire-and-forget message to another session or broadcast.

Args: from_sid: Your session ID. to_sid: Target session ID, or 'all' for broadcast. text: Message content.

Ask a question to another session (they see it on next heartbeat).

Args: from_sid: Your session ID. to_sid: Target session ID. question: The question text. Returns: Question ID (qid) for checking the answer later.

Answer a pending question from another session.

Args: qid: The question ID shown in heartbeat output. answer: Your response.

Check if a question has been answered.

Args: qid: The question ID from nexo_ask.

Check reminders and followups.

Args: filter: 'due' (vencidos/hoy), 'all' (todos activos), 'followups' (solo NEXO followups)

Generate the NEXO operations center menu with alerts and active sessions.

Shows: date, due alerts, all menu items by category, active sessions. Uses box-drawing characters for formatting.

Create a new reminder for the user.

Args: id: Unique ID starting with 'R' (e.g., R90). description: What needs to be done. date: Target date YYYY-MM-DD (optional). category: One of: decisiones, tareas, esperando, ideas, general.

Update fields of an existing reminder. Only non-empty fields are changed.

Args: id: Reminder ID (e.g., R87). description: New description (optional). date: New date YYYY-MM-DD (optional). status: New status (optional). category: New category (optional).

Mark a reminder as COMPLETADO with today's date.

Args: id: Reminder ID (e.g., R87).

Delete a reminder permanently.

Args: id: Reminder ID (e.g., R87).

Create a new NEXO followup (autonomous task).

Args: id: Unique ID starting with 'NF' (e.g., NF-MCP2). description: What to verify/do. date: Target date YYYY-MM-DD (optional). verification: How to verify completion (optional). reasoning: WHY this followup exists — what decision/context led to it (optional). recurrence: Auto-regenerate pattern (optional). Formats: 'weekly:monday', 'monthly:1', 'monthly:15', 'quarterly'. When completed, a new followup is auto-created with the next date. The completed one is archived with date suffix.

Update fields of an existing followup. Only non-empty fields are changed.

Args: id: Followup ID (e.g., NF45). description: New description (optional). date: New date YYYY-MM-DD (optional). verification: New verification text (optional). status: New status (optional).

Mark a followup as COMPLETADO. Appends result to verification field.

Args: id: Followup ID (e.g., NF45). result: What was found/done (optional).

Delete a followup permanently.

Args: id: Followup ID (e.g., NF45).

Add a new learning (resolved error, pattern, gotcha).

Args: category: One of: general, code, infrastructure, api, database, security, deployment, testing, performance, ux. title: Short title for the learning. content: Full description with context and solution. reasoning: WHY this matters — what led to discovering this (optional).

Search learnings by keyword. Searches title and content.

Args: query: Search term. category: Filter by category (optional).

Update a learning entry. Only non-empty fields are changed.

Args: id: Learning ID number. title: New title (optional). content: New content (optional). category: New category (optional).

Delete a learning entry.

Args: id: Learning ID number.

List all learnings, grouped by category.

Args: category: Filter by category (optional). If empty, shows all grouped.

Force full rebuild of the FTS5 search index. Use after bulk changes or if search seems stale.

Register a new directory for FTS5 search indexing. Survives restarts.

Args: path: Absolute path to directory (supports ~). dir_type: 'code' for source files, 'md' for markdown docs. patterns: Comma-separated glob patterns (only for code type). notes: Description of what this directory contains.

Remove a directory from FTS5 indexing and clean up its entries.

Args: path: Path to directory to remove.

List all directories being indexed by FTS5 (builtin + dynamic).

Get credential value(s) for a service.

Args: service: Service name (e.g., google-ads, meta-ads, shopify). key: Specific key (optional). If empty, returns all for the service.

Store a new credential.

Args: service: Service name (e.g., google-ads, cloudflare). key: Key name (e.g., api_key, token, ssh). value: The secret value. notes: Description or context (optional).

Update a credential's value and/or notes.

Args: service: Service name. key: Key name. value: New value (optional). notes: New notes (optional).

Delete credential(s). If no key, deletes all for the service.

Args: service: Service name. key: Specific key (optional). If empty, deletes ALL for service.

List credentials (names and notes only, no values).

Args: service: Filter by service (optional). If empty, shows all.

Record that an operational task was executed.

Args: task_num: Task number from the checklist (e.g., '7', '7b'). task_name: Task name (e.g., 'Google Ads'). notes: Execution summary (optional). reasoning: WHY this task was executed now — what triggered it (optional).

Show execution history for operational tasks.

Args: task_num: Filter by task number (optional). days: How many days back to show (default 30).

Check which operational tasks are overdue based on their frequency.

Compares last execution date vs configured frequency. Returns overdue tasks or 'all tasks up to date'.

Load or reload a plugin from the plugins/ directory.

Args: filename: Plugin filename (e.g., 'entities.py').

List all loaded plugins and their tools.

Remove a plugin: unregister its tools and delete the file.

Args: filename: Plugin filename (e.g., 'entities.py').