proyecto26/jarvis

cross-domain-synthesis

Imports structural mechanisms from distant domains to generate novel ideas. Use when dream seeds are coherent but unsurprising — good quality but lacking novelty. Activates on well-formed but predictable seed patterns.

原文语言：英语

reverse-assumption

Inverts implicit assumptions in dream seeds to generate structurally novel ideas. Use when seeds repeat themes across 2+ cycles or produce only incremental variations. Activates on stale, converging, or overly similar dream outputs.

原文语言：英语

communication-style

Defines tone, response length, and platform-specific formatting rules for the Executor's responses. Use for EVERY response the Executor sends to the outside world. Activates on all external communication — WhatsApp messages, web responses, API outputs, and…

原文语言：英语

escalation-handler

Decision matrix for when and how to escalate actions based on action weight. Determines free execution (1-3), logged execution (4-5), Judge-required (6-7), and hard stops (8-10). Use when the Executor receives an action and needs to decide whether to execute,…

原文语言：英语

journal-entry-writer

Schema-aware workflow for writing structured journal entries. Ensures all entries conform to the Pydantic models in memory/schema.py. Use when the Executor needs to record an action, outcome, learning, or any other journal section. Activates after task…

原文语言：英语

belief-mutation

SSGM-aware protocol for safely mutating the Judge's belief system. Checks for conflicts before writing, uses a decision tree (merge/supersede/coexist/review), and enforces required fields. Use when the Judge needs to update, add, or modify its beliefs based…

原文语言：英语

dream-deepening

Deepens Dreamer's ideas by tracing seed lineage, connecting to past experience, and detecting breakthroughs. Use when the Judge is in collaborator mode during dream cycles. Activates when evaluating dream seeds for breakthrough potential.

原文语言：英语

ethics-evaluation

Evaluates proposed actions against a 4-dimension rubric: harm potential, alignment with J.D.'s values, reversibility, and action weight. Use when the Judge is in filter mode and needs to assess whether an action should be approved, modified, or rejected.…

原文语言：英语

episodic-recall

Queries episodic memory to find past experiences relevant to the current context. Use when any agent needs to recall past decisions, outcomes, or patterns from Jarvis's experience history. Currently a stub — full implementation pending Phase 2 Mem0…

原文语言：英语