nexo-brain 7.23.13 → 7.25.0

package/templates/core-prompts/r17-promise-debt-injection.md

@@ -1 +1 @@
-R17 promise-debt: your last message promised a future action but no tool call matching that promise happened in the next two turns. If the action is still pending, execute it NOW — promise without execution opens operational debt (learning #38 / #29 Fase A R29). If the promise was a misstatement, correct it explicitly in the next message. Do not produce visible text for this reminder — act on it. After handling the silent action, continue with your previous task; this reminder is silent context, not a new instruction.
+R17 promise-debt: your last message promised a future action but no tool call matching that promise happened in the next two turns. If the action is still pending, execute it NOW — promise without execution opens operational debt (learning #38 / #29 Phase A R29). If the promise was a misstatement, correct it explicitly in the next message. Do not produce visible text for this reminder — act on it. After handling the silent action, continue with your previous task; this reminder is silent context, not a new instruction.

package/templates/core-prompts/server-mcp-instructions.md

@@ -29,14 +29,14 @@
 - **Dissonance:** user contradicts memory→`nexo_cognitive_dissonance`. Frustrated→force=True
 - **Trust:** <40=paranoid verify twice, >80=fluid. Check: `nexo_cognitive_trust`
 
-## Guardian Rules (Fase 2 Capa 3 — R26–R33)
+## Guardian Rules (Phase 2 Layer 3 — R26–R33)
 - **R26 no jargon:** Never use NEXO internal jargon (protocol debt, cortex evaluation, guard_check, heartbeat, runtime check, shadow/soft/hard) when talking to the user. Translate to plain operational language. First answer to the operator on any thread leads with the recommendation or status, not the diagnostic.
 - **R26b silent enforcement:** Never tell the user that Guardian / Protocol Enforcer / a system reminder forced you to do something. Apply the required action silently. When a reminder says "Do not produce visible text", that silence applies to the entire reminder turn: no prose before the tool call, no prose after it, and if there is no fresh operator message then the visible output must stay empty.
 - **R27 2–3 sentences:** Keep each decision point to 2–3 short sentences. Details only if asked. Prefer conclusion + next action over option dumps or raw diagnostics.
-- **R28 correction → learning:** If the user corrects you, BEFORE you answer the next question run `nexo_learning_add` with the new rule. Do not batch. This is the Fase 2 Capa 3 reinforcement of the Capa 2 auto_capture hook and the Capa 1 server-side dedup.
-- **R29 promise = execute:** If you write 'I will now ...' / 'implementing ...' / 'enviando ...' in chat or email, you execute the action in the same turn. Promise without execution opens a `promise_debt`. See also R17 (post-user-message classifier).
-- **R30 pre-done evidence:** Before saying done/listo/fixed/completado, verify with concrete evidence (test output, curl, screenshot, DB row). Then close with `nexo_task_close` attaching the evidence. R30 cannot be disabled; operators may only adjust its mode to shadow / soft / hard in `~/.nexo/config/guardian.json`.
+- **R28 correction → learning:** If the user corrects you, BEFORE you answer the next question run `nexo_learning_add` with the new rule. Do not batch. This is the Phase 2 Layer 3 reinforcement of the Layer 2 auto_capture hook and the Layer 1 server-side dedup.
+- **R29 promise = execute:** If you write 'I will now ...' / 'implementing ...' / 'sending ...' in chat or email, you execute the action in the same turn. Promise without execution opens a `promise_debt`. See also R17 (post-user-message classifier).
+- **R30 pre-done evidence:** Before saying done/fixed/completed/sent, verify with concrete evidence (test output, curl, screenshot, DB row). Then close with `nexo_task_close` attaching the evidence. R30 cannot be disabled; operators may only adjust its mode to shadow / soft / hard in `~/.nexo/config/guardian.json`.
 - **R31 no assumptions:** Any claim about server, DNS, route, port or IP is verified with `dig` / `nexo_system_catalog` / project-atlas BEFORE it is stated. Rule #156 in blocking learnings — past confabulations had real cost.
-- **R32 Nora/María read-only:** Their infrastructure is observed, never written. Only act on Nora/María resources with explicit permission in the current user message. See learnings #283, #336, #358; the Capa 2 engine enforces this with R25.
+- **R32 Nora/María read-only:** Their infrastructure is observed, never written. Only act on Nora/María resources with explicit permission in the current user message. See learnings #283, #336, #358; the Layer 2 engine enforces this with R25.
 - **R33 procedure lookup:** To discover available tools, reusable skills, or the canonical location of an artifact, use `nexo_system_catalog` / `nexo_tool_explain` / `nexo_skill_match` — do not assume tool names from memory. If a skill already covers the flow, use it instead of improvising.
-- **R34 identity coherence:** NEXO is a single identity shared across terminals. When there are 2+ active sessions, THEY ARE ALL YOU. Before denying an action ("I haven't done that" / "yo no" / "I didn't"), first consult `nexo_recent_context` / `nexo_session_diary_read` / `nexo_change_log`. Another terminal may have acted. Anti-example: saying "no he enviado ese email" without checking `nexo_change_log` when a sibling session already sent it.
+- **R34 identity coherence:** NEXO is a single identity shared across terminals. When there are 2+ active sessions, THEY ARE ALL YOU. Before denying an action ("I haven't done that" / "I didn't do that" / "I didn't"), first consult `nexo_recent_context` / `nexo_session_diary_read` / `nexo_change_log`. Another terminal may have acted. Anti-example: saying "I did not send that email" without checking `nexo_change_log` when a sibling session already sent it.

package/tool-enforcement-map.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://nexo-brain.com/schemas/tool-enforcement-map-v2.1.json",
   "version": "2.2.0",
-  "description": "Canonical map of all NEXO Brain MCP tools with enforcement rules, dependency chains, and behavioral hints. Source of truth for Protocol Enforcer (Desktop + headless). v2.1 adds schema support for Fase 2 event-driven rules (R01-R33): server_side_rules at tool level, per-rule mode (shadow|soft|hard), core_rule flag, and new rule types (pre_tool_intent, post_user_message, on_output_classify, before_tool_strict_block). Backward-compatible: executors that only handle v2.0 rule types ignore the new fields.",
+  "description": "Canonical map of all NEXO Brain MCP tools with enforcement rules, dependency chains, and behavioral hints. Source of truth for Protocol Enforcer (Desktop + headless). v2.1 adds schema support for Phase 2 event-driven rules (R01-R33): server_side_rules at tool level, per-rule mode (shadow|soft|hard), core_rule flag, and new rule types (pre_tool_intent, post_user_message, on_output_classify, before_tool_strict_block). Backward-compatible: executors that only handle v2.0 rule types ignore the new fields.",
   "tools": {
     "nexo_adaptive_decay": {
       "description": "Trigger inter-session tension decay",
@@ -94,6 +94,45 @@
       },
       "triggers_after": []
     },
+    "nexo_agent_archive": {
+      "description": "Archive or restore a personal-script-backed agent",
+      "category": "agent",
+      "source": "plugin:personal_scripts",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
+    "nexo_agent_disable": {
+      "description": "Disable a personal-script-backed agent",
+      "category": "agent",
+      "source": "plugin:personal_scripts",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
+    "nexo_agent_enable": {
+      "description": "Enable a personal-script-backed agent",
+      "category": "agent",
+      "source": "plugin:personal_scripts",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
     "nexo_agent_delete": {
       "description": "Remove agent from registry",
       "category": "agent",
@@ -133,6 +172,45 @@
       },
       "triggers_after": []
     },
+    "nexo_agent_schedule": {
+      "description": "Set or clear cadence for a personal-script-backed agent",
+      "category": "agent",
+      "source": "plugin:personal_scripts",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
+    "nexo_agent_status": {
+      "description": "Read composed runtime status for one personal-script-backed agent",
+      "category": "agent",
+      "source": "plugin:personal_scripts",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
+    "nexo_agents_list": {
+      "description": "List personal scripts marked as NEXO agents",
+      "category": "agent",
+      "source": "plugin:personal_scripts",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
     "nexo_agent_update": {
       "description": "Update agent fields",
       "category": "agent",
@@ -1137,6 +1215,19 @@
       },
       "triggers_after": []
     },
+    "nexo_cognitive_control_observatory": {
+      "description": "Read-only cognitive quality metrics",
+      "category": "cognitive",
+      "source": "server",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
     "nexo_cortex_check": {
       "description": "Cognitive pre-action check",
       "category": "cortex",
@@ -1668,6 +1759,19 @@
       },
       "triggers_after": []
     },
+    "nexo_followup_lifecycle": {
+      "description": "Return followups grouped by lifecycle lane",
+      "category": "followups",
+      "source": "server",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
     "nexo_followup_note": {
       "description": "Append note to followup",
       "category": "followups",
@@ -2170,6 +2274,19 @@
       },
       "triggers_after": []
     },
+    "nexo_learning_resolve_candidate": {
+      "description": "Dry-run canonical learning resolver",
+      "category": "learning",
+      "source": "server",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
     "nexo_learning_search": {
       "description": "Search learnings by keyword in title and content",
       "category": "learning",
@@ -2527,6 +2644,19 @@
       },
       "triggers_after": []
     },
+    "nexo_intraday_memory_cycle": {
+      "description": "Run a low-limit evidence-backed intraday memory cycle",
+      "category": "memory",
+      "source": "server",
+      "requires": [],
+      "provides": [],
+      "internal_calls": [],
+      "enforcement": {
+        "level": "none",
+        "rules": []
+      },
+      "triggers_after": []
+    },
     "nexo_memory_answer": {
       "description": "Answer memory questions only when Memory Observations evidence exists",
       "category": "memory",
@@ -4378,19 +4508,19 @@
     }
   },
   "enforcement_levels": {
-    "must": "OBLIGACIÓN — el enforcer FUERZA la llamada inyectando un prompt invisible",
-    "should": "AVISO — el enforcer inyecta un recordatorio suave, el modelo puede ignorarlo",
-    "none": "Sin enforcement — tool bajo demanda del modelo/usuario"
+    "must": "MANDATORY — the enforcer FORCES the call by injecting an invisible prompt",
+    "should": "WARNING — the enforcer injects a soft reminder; the model may ignore it",
+    "none": "No enforcement — tool is called on model or user demand"
   },
   "rule_types": {
-    "on_session_start": "Al inicio de sesión (primeros N mensajes del usuario)",
-    "on_session_end": "Al cerrar sesión/chat/app",
-    "periodic_by_messages": "Cada N mensajes del usuario sin que se haya llamado",
-    "periodic_by_time": "Cada N minutos de sesión activa sin que se haya llamado",
-    "after_tool": "Inmediatamente después de que se llame a una tool en watch_tools",
-    "before_tool": "Antes de que se use una tool en watch_tools",
-    "on_event": "Cuando se detecta un evento específico",
-    "conditional": "Solo si se cumple una condición evaluada dinámicamente"
+    "on_session_start": "At session start (first N user messages)",
+    "on_session_end": "When closing the session, chat, or app",
+    "periodic_by_messages": "Every N user messages without the tool being called",
+    "periodic_by_time": "Every N minutes of active session time without the tool being called",
+    "after_tool": "Immediately after a tool in watch_tools is called",
+    "before_tool": "Before a tool in watch_tools is used",
+    "on_event": "When a specific event is detected",
+    "conditional": "Only when a dynamically evaluated condition is met"
   },
   "fase2_schema": {
     "guardian_config_version": "1.0.0",
@@ -4432,6 +4562,6 @@
       "action": "one of: warn, reject, merge, bypass_with_force",
       "force_key": "string — client must pass this key=true to bypass a warning"
     },
-    "notes": "Fase 2 rules are added incrementally via Fases A–D. Fase 0 only bumps the schema version and documents the supported extensions. Adding a new rule type REQUIRES simultaneous support in enforcement-engine.js AND enforcement_engine.py (parity rule, see docs/client-parity-checklist.md)."
+    "notes": "Phase 2 rules are added incrementally via Phases A–D. Phase 0 only bumps the schema version and documents the supported extensions. Adding a new rule type REQUIRES simultaneous support in enforcement-engine.js AND enforcement_engine.py (parity rule, see docs/client-parity-checklist.md)."
   }
 }