菜单
Biologically-inspired cognitive memory for AI agents. Panksepp affects, Free Energy consolidation, VSA binding, sparse retrieval, temporal prediction, metacognition. Website - https://nima-core.ai
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/duclm1x1/Dive-Ai --skill nima-core命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Persistent memory system for AI agents following Model Context Protocol (MCP). Use for storing long-term memories across sessions, semantic search of past knowledge, building knowledge graphs, auto-injecting context, deduplicating memories, syncing to cloud storage. Essential for agents that need to remember decisions, solutions, preferences, and learned patterns over time.
Persistent memory system for AI agents following Model Context Protocol (MCP). Use for storing long-term memories across sessions, semantic search of past knowledge, building knowledge graphs, auto-injecting context, deduplicating memories, syncing to cloud storage. Essential for agents that need to remember decisions, solutions, preferences, and learned patterns over time.
Master REST and GraphQL API design principles to build intuitive, scalable, and maintainable APIs that delight developers. Use when designing new APIs, reviewing API specifications, or establishing API design standards.
正在显示 SKILL.md
基于 SOC 职业分类
| name | nima-core |
| description | Biologically-inspired cognitive memory for AI agents. Panksepp affects, Free Energy consolidation, VSA binding, sparse retrieval, temporal prediction, metacognition. Website - https://nima-core.ai |
| version | 1.1.3 |
| metadata | {"openclaw":{"emoji":"🧠","requires":{"bins":["python3"]}}} |
Plug-and-play cognitive memory architecture for AI agents.
Website: https://nima-core.ai
GitHub: https://github.com/lilubot/nima-core
pip install nima-core
# Auto-setup (detects OpenClaw, installs hooks, configures everything)
nima-core
openclaw gateway restart
openclaw hooks install /path/to/nima-core
openclaw hooks enable nima-bootstrap
openclaw hooks enable nima-recall
openclaw gateway restart
from nima_core import NimaCore
nima = NimaCore(
name="MyBot",
important_people={"Alice": 1.5, "Bob": 1.3}, # These people's memories get priority
)
nima.experience("Alice asked about the project", who="Alice", importance=0.7)
results = nima.recall("project")
Configure which people, emotions, and topics matter most:
from nima_core.services.heartbeat import NimaHeartbeat, SmartConsolidation
smart = SmartConsolidation(
important_people={"Alice": 1.5, "Bob": 1.3}, # Weight multipliers
emotion_words={"love", "excited", "proud", "worried"}, # Force-consolidate
importance_markers={"family", "milestone", "decision"}, # Force-consolidate
noise_patterns=["system exec", "heartbeat_ok"], # Skip these
)
heartbeat = NimaHeartbeat(nima, message_source=my_source, smart_consolidation=smart)
heartbeat.start_background()
Memories from important people get boosted. Emotional content is always kept. Noise is filtered out.
All V2 components are enabled by default (v1.1.0+). No configuration needed.
To disable: export NIMA_V2_ALL=false
nima.experience(content, who, importance) — Process through affect → binding → FE pipelinenima.recall(query, top_k) — Semantic memory searchnima.capture(who, what, importance) — Explicit memory capture (bypasses FE gate)nima.synthesize(insight, domain, sparked_by, importance) — Lightweight insight capture (280 char max)nima.dream(hours) — Run consolidation (schema extraction)nima.status() — System statusnima.introspect() — Metacognitive self-reflectionMETACOGNITIVE — Self-model, 4-chunk WM, strange loops
SEMANTIC — Hyperbolic embeddings, concept hierarchies
EPISODIC — VSA + Holographic storage, sparse retrieval
CONSOLIDATION — Free Energy decisions, schema extraction
BINDING — VSA circular convolution, role-filler composition
AFFECTIVE CORE — Panksepp's 7 affects (SEEKING, RAGE, FEAR, LUST, CARE, PANIC, PLAY)
| Variable | Default | Description |
|---|---|---|
NIMA_DATA_DIR | ./nima_data | Memory storage path |
NIMA_MODELS_DIR | ./models | Model files path |
NIMA_V2_ALL | true | Full cognitive stack (affects, binding, FE, etc.) |
NIMA_SPARSE_RETRIEVAL | true | Two-stage sparse index |
NIMA_PROJECTION | true | 384D → 50KD projection |
README.md — Full documentation with all settingsnima_core/config/nima_config.py — All feature flags.env.example — Environment variable template