{ "schemaVersion": "1.0", "item": { "slug": "nima-core", "name": "NIMA Core", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/dmdorta1111/nima-core", "canonicalUrl": "https://clawhub.ai/dmdorta1111/nima-core", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/nima-core", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=nima-core", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "CHANGELOG.md", "INSTALL.md", "QUICKSTART.md", "README.md", "SKILL.md", "doctor.sh" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/nima-core" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/nima-core", "agentPageUrl": "https://openagent3.xyz/skills/nima-core/agent", "manifestUrl": "https://openagent3.xyz/skills/nima-core/agent.json", "briefUrl": "https://openagent3.xyz/skills/nima-core/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "NIMA Core 3.2", "body": "Noosphere Integrated Memory Architecture — A complete cognitive stack for AI agents: persistent memory, emotional intelligence, dream consolidation, hive mind, and precognitive recall.\n\nWebsite: https://nima-core.ai · GitHub: https://github.com/lilubot/nima-core" }, { "title": "Quick Start", "body": "pip install nima-core && nima-core\n\nYour bot now has persistent memory. Zero config needed." }, { "title": "Complete Cognitive Stack", "body": "NIMA evolved from a memory plugin into a full cognitive architecture:\n\nModuleWhat It DoesVersionMemory Capture3-layer capture (input/contemplation/output), 4-phase noise filteringv2.0Semantic RecallVector + text hybrid search, ecology scoring, token-budgeted injectionv2.0Dynamic AffectPanksepp 7-affect emotional state (SEEKING, RAGE, FEAR, LUST, CARE, PANIC, PLAY)v2.1VADER AnalyzerContextual sentiment — caps boost, negation, idioms, degree modifiersv2.2Memory PrunerLLM distillation of old conversations → semantic gists, 30-day suppression limbov2.3Dream ConsolidationNightly synthesis — extracts insights and patterns from episodic memoryv2.4Hive MindMulti-agent memory sharing via shared DB + optional Redis pub/subv2.5PrecognitionTemporal pattern mining → predictive memory pre-loadingv2.5Lucid MomentsSpontaneous surfacing of emotionally-resonant memoriesv2.5Darwinian MemoryClusters similar memories, ghosts duplicates via cosine + LLM verificationv3.0InstallerOne-command setup — LadybugDB, hooks, directories, embedder configv3.0" }, { "title": "v3.0 Highlights", "body": "All cognitive modules unified under a single package\nInstaller (install.sh) for zero-friction setup\nAll OpenClaw hooks bundled and ready to drop in\nREADME rewritten, all versions aligned to 3.0.4" }, { "title": "Architecture", "body": "OPENCLAW HOOKS\n├── nima-memory/ Capture hook (3-layer, 4-phase noise filter)\n│ ├── index.js Hook entry point\n│ ├── ladybug_store.py LadybugDB storage backend\n│ ├── embeddings.py Multi-provider embedding (Voyage/OpenAI/Ollama/local)\n│ ├── backfill.py Historical transcript import\n│ └── health_check.py DB integrity checks\n├── nima-recall-live/ Recall hook (before_agent_start)\n│ ├── lazy_recall.py Current recall engine\n│ └── ladybug_recall.py LadybugDB-native recall\n├── nima-affect/ Affect hook (message_received)\n│ ├── vader-affect.js VADER sentiment analyzer\n│ └── emotion-lexicon.js Emotion keyword lexicon\n└── shared/ Resilient wrappers, error handling\n\nPYTHON CORE (nima_core/)\n├── cognition/\n│ ├── dynamic_affect.py Panksepp 7-affect system\n│ ├── emotion_detection.py Text emotion extraction\n│ ├── affect_correlation.py Cross-affect analysis\n│ ├── affect_history.py Temporal affect tracking\n│ ├── affect_interactions.py Affect coupling dynamics\n│ ├── archetypes.py Personality baselines (Guardian, Explorer, etc.)\n│ ├── personality_profiles.py JSON personality configs\n│ └── response_modulator_v2.py Affect → response modulation\n├── dream_consolidation.py Nightly memory synthesis engine\n├── memory_pruner.py Episodic distillation + suppression\n├── hive_mind.py Multi-agent memory sharing\n├── precognition.py Temporal pattern mining\n├── lucid_moments.py Spontaneous memory surfacing\n├── connection_pool.py SQLite pool (WAL, thread-safe)\n├── logging_config.py Singleton logger\n└── metrics.py Thread-safe counters/timings" }, { "title": "Privacy & Permissions", "body": "✅ All data stored locally in ~/.nima/\n✅ Default: local embeddings = zero external calls\n✅ No NIMA-owned servers, no proprietary tracking, no analytics sent to external services\n⚠️ Opt-in networking: HiveMind (Redis pub/sub), Precognition (LLM endpoints), LadybugDB migrations — see Optional Features below\n🔒 Embedding API calls only when explicitly enabling (VOYAGE_API_KEY, OPENAI_API_KEY, etc.)" }, { "title": "Optional Features with Network Access", "body": "FeatureEnv VarNetwork Calls ToDefaultCloud embeddingsNIMA_EMBEDDER=voyagevoyage.aiOffCloud embeddingsNIMA_EMBEDDER=openaiopenai.comOffMemory prunerANTHROPIC_API_KEY setanthropic.comOffOllama embeddingsNIMA_EMBEDDER=ollamalocalhost:11434OffHiveMindHIVE_ENABLED=trueRedis pub/subOffPrecognitionUsing external LLMConfigured endpointOff" }, { "title": "What Gets Installed", "body": "ComponentLocationPurposePython core (nima_core/)~/.nima/Memory, affect, cognitionOpenClaw hooks~/.openclaw/extensions/nima-*/Capture, recall, affectSQLite database~/.nima/memory/graph.sqlitePersistent storageLogs~/.nima/logs/Debug logs (optional)" }, { "title": "Credential Handling", "body": "Env VarRequired?Network Calls?PurposeNIMA_EMBEDDER=localNo❌Default — offline embeddingsVOYAGE_API_KEYOnly if using Voyage✅ voyage.aiCloud embeddingsOPENAI_API_KEYOnly if using OpenAI✅ openai.comCloud embeddingsANTHROPIC_API_KEYOnly if using pruner✅ anthropic.comMemory distillationNIMA_OLLAMA_MODELOnly if using Ollama❌ (localhost)Local GPU embeddings\n\nRecommendation: Start with NIMA_EMBEDDER=local (default). Only enable cloud providers when you need better embedding quality." }, { "title": "Safety Features", "body": "Input filtering — System messages, heartbeats, and duplicates are filtered before capture\nFTS5 injection prevention — Parameterized queries prevent SQL injection\nPath traversal protection — All file paths are sanitized\nTemp file cleanup — Automatic cleanup of temporary files\nAPI timeouts — Network calls have reasonable timeouts (30s Voyage, 10s local)" }, { "title": "Best Practices", "body": "Review before installing — Inspect install.sh and hook files before running\nBackup config — Backup ~/.openclaw/openclaw.json before adding hooks\nDon't run as root — Installation writes to user home directories\nUse containerized envs — Test in a VM or container first if unsure\nRotate API keys — If using cloud embeddings, rotate keys periodically\nMonitor logs — Check ~/.nima/logs/ for suspicious activity" }, { "title": "Data Locations", "body": "~/.nima/\n├── memory/\n│ ├── graph.sqlite # SQLite backend (default)\n│ ├── ladybug.lbug # LadybugDB backend (optional)\n│ ├── embedding_cache.db # Cached embeddings\n│ └── embedding_index.npy# Vector index\n├── affect/\n│ └── affect_state.json # Current emotional state\n└── logs/ # Debug logs (if enabled)\n\n~/.openclaw/extensions/\n├── nima-memory/ # Capture hook\n├── nima-recall-live/ # Recall hook\n└── nima-affect/ # Affect hook\n\nControls:\n\n{\n \"plugins\": {\n \"entries\": {\n \"nima-memory\": {\n \"skip_subagents\": true,\n \"skip_heartbeats\": true,\n \"noise_filtering\": { \"filter_system_noise\": true }\n }\n }\n }\n}" }, { "title": "Embedding Providers", "body": "ProviderSetupDimsCostLocal (default)NIMA_EMBEDDER=local384FreeVoyage AINIMA_EMBEDDER=voyage + VOYAGE_API_KEY1024$0.12/1M tokOpenAINIMA_EMBEDDER=openai + OPENAI_API_KEY1536$0.13/1M tokOllamaNIMA_EMBEDDER=ollama + NIMA_OLLAMA_MODEL768Free" }, { "title": "Database Backend", "body": "SQLite (default)LadybugDB (recommended)Text Search31ms9ms (3.4x faster)Vector SearchExternalNative HNSW (18ms)Graph QueriesSQL JOINsNative CypherDB Size~91 MB~50 MB (44% smaller)\n\nUpgrade: pip install real-ladybug && python -c \"from nima_core.storage import migrate; migrate()\"" }, { "title": "All Environment Variables", "body": "# Embedding (default: local)\nNIMA_EMBEDDER=local|voyage|openai|ollama\nVOYAGE_API_KEY=pa-xxx\nOPENAI_API_KEY=sk-xxx\nNIMA_OLLAMA_MODEL=nomic-embed-text\n\n# Data paths\nNIMA_DATA_DIR=~/.nima\nNIMA_DB_PATH=~/.nima/memory/ladybug.lbug\n\n# Memory pruner\nNIMA_DISTILL_MODEL=claude-haiku-4-5\nANTHROPIC_API_KEY=sk-ant-xxx\n\n# Logging\nNIMA_LOG_LEVEL=INFO\nNIMA_DEBUG_RECALL=1" }, { "title": "Hooks", "body": "HookFiresDoesnima-memoryAfter saveCaptures 3 layers → filters noise → stores in graph DBnima-recall-liveBefore LLMSearches memories → scores by ecology → injects as context (3000 token budget)nima-affectOn messageVADER sentiment → Panksepp 7-affect state → archetype modulation" }, { "title": "Installation", "body": "./install.sh\nopenclaw gateway restart\n\nOr manual:\n\ncp -r openclaw_hooks/nima-memory ~/.openclaw/extensions/\ncp -r openclaw_hooks/nima-recall-live ~/.openclaw/extensions/\ncp -r openclaw_hooks/nima-affect ~/.openclaw/extensions/" }, { "title": "Dream Consolidation", "body": "Nightly synthesis extracts insights and patterns from episodic memory:\n\npython -m nima_core.dream_consolidation\n# Or schedule via OpenClaw cron at 2 AM" }, { "title": "Memory Pruner", "body": "Distills old conversations into semantic gists, suppresses raw noise:\n\npython -m nima_core.memory_pruner --min-age 14 --live\npython -m nima_core.memory_pruner --restore 12345 # undo within 30 days" }, { "title": "Hive Mind", "body": "Multi-agent memory sharing:\n\nfrom nima_core import HiveMind\nhive = HiveMind(db_path=\"~/.nima/memory/ladybug.lbug\")\ncontext = hive.build_agent_context(\"research task\", max_memories=8)\nhive.capture_agent_result(\"agent-1\", \"result summary\", \"model-name\")" }, { "title": "Precognition", "body": "Temporal pattern mining → predictive memory pre-loading:\n\nfrom nima_core import NimaPrecognition\nprecog = NimaPrecognition(db_path=\"~/.nima/memory/ladybug.lbug\")\nprecog.run_mining_cycle()" }, { "title": "Lucid Moments", "body": "Spontaneous surfacing of emotionally-resonant memories (with safety: trauma filtering, quiet hours, daily caps):\n\nfrom nima_core import LucidMoments\nlucid = LucidMoments(db_path=\"~/.nima/memory/ladybug.lbug\")\nmoment = lucid.surface_moment()" }, { "title": "Affect System", "body": "Panksepp 7-affect emotional intelligence with personality archetypes:\n\nfrom nima_core import DynamicAffectSystem\naffect = DynamicAffectSystem(identity_name=\"my_bot\", baseline=\"guardian\")\nstate = affect.process_input(\"I'm excited about this!\")\n# Archetypes: guardian, explorer, trickster, empath, sage" }, { "title": "API", "body": "from nima_core import (\n DynamicAffectSystem,\n get_affect_system,\n HiveMind,\n NimaPrecognition,\n LucidMoments,\n)\n\n# Affect (thread-safe singleton)\naffect = get_affect_system(identity_name=\"lilu\")\nstate = affect.process_input(\"Hello!\")\n\n# Hive Mind\nhive = HiveMind()\ncontext = hive.build_agent_context(\"task description\")\n\n# Precognition\nprecog = NimaPrecognition()\nprecog.run_mining_cycle()\n\n# Lucid Moments\nlucid = LucidMoments()\nmoment = lucid.surface_moment()" }, { "title": "Changelog", "body": "See CHANGELOG.md for full version history." }, { "title": "Recent Releases", "body": "v3.0.4 (Feb 23, 2026) — Darwinian memory engine, new CLIs, installer, bug fixes\nv2.5.0 (Feb 21, 2026) — Hive Mind, Precognition, Lucid Moments\nv2.4.0 (Feb 20, 2026) — Dream Consolidation engine\nv2.3.0 (Feb 19, 2026) — Memory Pruner, connection pool, Ollama support\nv2.2.0 (Feb 19, 2026) — VADER Affect, 4-phase noise remediation, ecology scoring\nv2.0.0 (Feb 13, 2026) — LadybugDB backend, security hardening, 348 tests" }, { "title": "License", "body": "MIT — free for any AI agent, commercial or personal." } ], "body": "NIMA Core 3.2\n\nNoosphere Integrated Memory Architecture — A complete cognitive stack for AI agents: persistent memory, emotional intelligence, dream consolidation, hive mind, and precognitive recall.\n\nWebsite: https://nima-core.ai · GitHub: https://github.com/lilubot/nima-core\n\nQuick Start\npip install nima-core && nima-core\n\n\nYour bot now has persistent memory. Zero config needed.\n\nWhat's New in v3.0\nComplete Cognitive Stack\n\nNIMA evolved from a memory plugin into a full cognitive architecture:\n\nModule\tWhat It Does\tVersion\nMemory Capture\t3-layer capture (input/contemplation/output), 4-phase noise filtering\tv2.0\nSemantic Recall\tVector + text hybrid search, ecology scoring, token-budgeted injection\tv2.0\nDynamic Affect\tPanksepp 7-affect emotional state (SEEKING, RAGE, FEAR, LUST, CARE, PANIC, PLAY)\tv2.1\nVADER Analyzer\tContextual sentiment — caps boost, negation, idioms, degree modifiers\tv2.2\nMemory Pruner\tLLM distillation of old conversations → semantic gists, 30-day suppression limbo\tv2.3\nDream Consolidation\tNightly synthesis — extracts insights and patterns from episodic memory\tv2.4\nHive Mind\tMulti-agent memory sharing via shared DB + optional Redis pub/sub\tv2.5\nPrecognition\tTemporal pattern mining → predictive memory pre-loading\tv2.5\nLucid Moments\tSpontaneous surfacing of emotionally-resonant memories\tv2.5\nDarwinian Memory\tClusters similar memories, ghosts duplicates via cosine + LLM verification\tv3.0\nInstaller\tOne-command setup — LadybugDB, hooks, directories, embedder config\tv3.0\nv3.0 Highlights\nAll cognitive modules unified under a single package\nInstaller (install.sh) for zero-friction setup\nAll OpenClaw hooks bundled and ready to drop in\nREADME rewritten, all versions aligned to 3.0.4\nArchitecture\nOPENCLAW HOOKS\n├── nima-memory/ Capture hook (3-layer, 4-phase noise filter)\n│ ├── index.js Hook entry point\n│ ├── ladybug_store.py LadybugDB storage backend\n│ ├── embeddings.py Multi-provider embedding (Voyage/OpenAI/Ollama/local)\n│ ├── backfill.py Historical transcript import\n│ └── health_check.py DB integrity checks\n├── nima-recall-live/ Recall hook (before_agent_start)\n│ ├── lazy_recall.py Current recall engine\n│ └── ladybug_recall.py LadybugDB-native recall\n├── nima-affect/ Affect hook (message_received)\n│ ├── vader-affect.js VADER sentiment analyzer\n│ └── emotion-lexicon.js Emotion keyword lexicon\n└── shared/ Resilient wrappers, error handling\n\nPYTHON CORE (nima_core/)\n├── cognition/\n│ ├── dynamic_affect.py Panksepp 7-affect system\n│ ├── emotion_detection.py Text emotion extraction\n│ ├── affect_correlation.py Cross-affect analysis\n│ ├── affect_history.py Temporal affect tracking\n│ ├── affect_interactions.py Affect coupling dynamics\n│ ├── archetypes.py Personality baselines (Guardian, Explorer, etc.)\n│ ├── personality_profiles.py JSON personality configs\n│ └── response_modulator_v2.py Affect → response modulation\n├── dream_consolidation.py Nightly memory synthesis engine\n├── memory_pruner.py Episodic distillation + suppression\n├── hive_mind.py Multi-agent memory sharing\n├── precognition.py Temporal pattern mining\n├── lucid_moments.py Spontaneous memory surfacing\n├── connection_pool.py SQLite pool (WAL, thread-safe)\n├── logging_config.py Singleton logger\n└── metrics.py Thread-safe counters/timings\n\nPrivacy & Permissions\n✅ All data stored locally in ~/.nima/\n✅ Default: local embeddings = zero external calls\n✅ No NIMA-owned servers, no proprietary tracking, no analytics sent to external services\n⚠️ Opt-in networking: HiveMind (Redis pub/sub), Precognition (LLM endpoints), LadybugDB migrations — see Optional Features below\n🔒 Embedding API calls only when explicitly enabling (VOYAGE_API_KEY, OPENAI_API_KEY, etc.)\nOptional Features with Network Access\nFeature\tEnv Var\tNetwork Calls To\tDefault\nCloud embeddings\tNIMA_EMBEDDER=voyage\tvoyage.ai\tOff\nCloud embeddings\tNIMA_EMBEDDER=openai\topenai.com\tOff\nMemory pruner\tANTHROPIC_API_KEY set\tanthropic.com\tOff\nOllama embeddings\tNIMA_EMBEDDER=ollama\tlocalhost:11434\tOff\nHiveMind\tHIVE_ENABLED=true\tRedis pub/sub\tOff\nPrecognition\tUsing external LLM\tConfigured endpoint\tOff\nSecurity\nWhat Gets Installed\nComponent\tLocation\tPurpose\nPython core (nima_core/)\t~/.nima/\tMemory, affect, cognition\nOpenClaw hooks\t~/.openclaw/extensions/nima-*/\tCapture, recall, affect\nSQLite database\t~/.nima/memory/graph.sqlite\tPersistent storage\nLogs\t~/.nima/logs/\tDebug logs (optional)\nCredential Handling\nEnv Var\tRequired?\tNetwork Calls?\tPurpose\nNIMA_EMBEDDER=local\tNo\t❌\tDefault — offline embeddings\nVOYAGE_API_KEY\tOnly if using Voyage\t✅ voyage.ai\tCloud embeddings\nOPENAI_API_KEY\tOnly if using OpenAI\t✅ openai.com\tCloud embeddings\nANTHROPIC_API_KEY\tOnly if using pruner\t✅ anthropic.com\tMemory distillation\nNIMA_OLLAMA_MODEL\tOnly if using Ollama\t❌ (localhost)\tLocal GPU embeddings\n\nRecommendation: Start with NIMA_EMBEDDER=local (default). Only enable cloud providers when you need better embedding quality.\n\nSafety Features\nInput filtering — System messages, heartbeats, and duplicates are filtered before capture\nFTS5 injection prevention — Parameterized queries prevent SQL injection\nPath traversal protection — All file paths are sanitized\nTemp file cleanup — Automatic cleanup of temporary files\nAPI timeouts — Network calls have reasonable timeouts (30s Voyage, 10s local)\nBest Practices\nReview before installing — Inspect install.sh and hook files before running\nBackup config — Backup ~/.openclaw/openclaw.json before adding hooks\nDon't run as root — Installation writes to user home directories\nUse containerized envs — Test in a VM or container first if unsure\nRotate API keys — If using cloud embeddings, rotate keys periodically\nMonitor logs — Check ~/.nima/logs/ for suspicious activity\nData Locations\n~/.nima/\n├── memory/\n│ ├── graph.sqlite # SQLite backend (default)\n│ ├── ladybug.lbug # LadybugDB backend (optional)\n│ ├── embedding_cache.db # Cached embeddings\n│ └── embedding_index.npy# Vector index\n├── affect/\n│ └── affect_state.json # Current emotional state\n└── logs/ # Debug logs (if enabled)\n\n~/.openclaw/extensions/\n├── nima-memory/ # Capture hook\n├── nima-recall-live/ # Recall hook\n└── nima-affect/ # Affect hook\n\n\nControls:\n\n{\n \"plugins\": {\n \"entries\": {\n \"nima-memory\": {\n \"skip_subagents\": true,\n \"skip_heartbeats\": true,\n \"noise_filtering\": { \"filter_system_noise\": true }\n }\n }\n }\n}\n\nConfiguration\nEmbedding Providers\nProvider\tSetup\tDims\tCost\nLocal (default)\tNIMA_EMBEDDER=local\t384\tFree\nVoyage AI\tNIMA_EMBEDDER=voyage + VOYAGE_API_KEY\t1024\t$0.12/1M tok\nOpenAI\tNIMA_EMBEDDER=openai + OPENAI_API_KEY\t1536\t$0.13/1M tok\nOllama\tNIMA_EMBEDDER=ollama + NIMA_OLLAMA_MODEL\t768\tFree\nDatabase Backend\n\tSQLite (default)\tLadybugDB (recommended)\nText Search\t31ms\t9ms (3.4x faster)\nVector Search\tExternal\tNative HNSW (18ms)\nGraph Queries\tSQL JOINs\tNative Cypher\nDB Size\t~91 MB\t~50 MB (44% smaller)\n\nUpgrade: pip install real-ladybug && python -c \"from nima_core.storage import migrate; migrate()\"\n\nAll Environment Variables\n# Embedding (default: local)\nNIMA_EMBEDDER=local|voyage|openai|ollama\nVOYAGE_API_KEY=pa-xxx\nOPENAI_API_KEY=sk-xxx\nNIMA_OLLAMA_MODEL=nomic-embed-text\n\n# Data paths\nNIMA_DATA_DIR=~/.nima\nNIMA_DB_PATH=~/.nima/memory/ladybug.lbug\n\n# Memory pruner\nNIMA_DISTILL_MODEL=claude-haiku-4-5\nANTHROPIC_API_KEY=sk-ant-xxx\n\n# Logging\nNIMA_LOG_LEVEL=INFO\nNIMA_DEBUG_RECALL=1\n\nHooks\nHook\tFires\tDoes\nnima-memory\tAfter save\tCaptures 3 layers → filters noise → stores in graph DB\nnima-recall-live\tBefore LLM\tSearches memories → scores by ecology → injects as context (3000 token budget)\nnima-affect\tOn message\tVADER sentiment → Panksepp 7-affect state → archetype modulation\nInstallation\n./install.sh\nopenclaw gateway restart\n\n\nOr manual:\n\ncp -r openclaw_hooks/nima-memory ~/.openclaw/extensions/\ncp -r openclaw_hooks/nima-recall-live ~/.openclaw/extensions/\ncp -r openclaw_hooks/nima-affect ~/.openclaw/extensions/\n\nAdvanced Features\nDream Consolidation\n\nNightly synthesis extracts insights and patterns from episodic memory:\n\npython -m nima_core.dream_consolidation\n# Or schedule via OpenClaw cron at 2 AM\n\nMemory Pruner\n\nDistills old conversations into semantic gists, suppresses raw noise:\n\npython -m nima_core.memory_pruner --min-age 14 --live\npython -m nima_core.memory_pruner --restore 12345 # undo within 30 days\n\nHive Mind\n\nMulti-agent memory sharing:\n\nfrom nima_core import HiveMind\nhive = HiveMind(db_path=\"~/.nima/memory/ladybug.lbug\")\ncontext = hive.build_agent_context(\"research task\", max_memories=8)\nhive.capture_agent_result(\"agent-1\", \"result summary\", \"model-name\")\n\nPrecognition\n\nTemporal pattern mining → predictive memory pre-loading:\n\nfrom nima_core import NimaPrecognition\nprecog = NimaPrecognition(db_path=\"~/.nima/memory/ladybug.lbug\")\nprecog.run_mining_cycle()\n\nLucid Moments\n\nSpontaneous surfacing of emotionally-resonant memories (with safety: trauma filtering, quiet hours, daily caps):\n\nfrom nima_core import LucidMoments\nlucid = LucidMoments(db_path=\"~/.nima/memory/ladybug.lbug\")\nmoment = lucid.surface_moment()\n\nAffect System\n\nPanksepp 7-affect emotional intelligence with personality archetypes:\n\nfrom nima_core import DynamicAffectSystem\naffect = DynamicAffectSystem(identity_name=\"my_bot\", baseline=\"guardian\")\nstate = affect.process_input(\"I'm excited about this!\")\n# Archetypes: guardian, explorer, trickster, empath, sage\n\nAPI\nfrom nima_core import (\n DynamicAffectSystem,\n get_affect_system,\n HiveMind,\n NimaPrecognition,\n LucidMoments,\n)\n\n# Affect (thread-safe singleton)\naffect = get_affect_system(identity_name=\"lilu\")\nstate = affect.process_input(\"Hello!\")\n\n# Hive Mind\nhive = HiveMind()\ncontext = hive.build_agent_context(\"task description\")\n\n# Precognition\nprecog = NimaPrecognition()\nprecog.run_mining_cycle()\n\n# Lucid Moments\nlucid = LucidMoments()\nmoment = lucid.surface_moment()\n\nChangelog\n\nSee CHANGELOG.md for full version history.\n\nRecent Releases\nv3.0.4 (Feb 23, 2026) — Darwinian memory engine, new CLIs, installer, bug fixes\nv2.5.0 (Feb 21, 2026) — Hive Mind, Precognition, Lucid Moments\nv2.4.0 (Feb 20, 2026) — Dream Consolidation engine\nv2.3.0 (Feb 19, 2026) — Memory Pruner, connection pool, Ollama support\nv2.2.0 (Feb 19, 2026) — VADER Affect, 4-phase noise remediation, ecology scoring\nv2.0.0 (Feb 13, 2026) — LadybugDB backend, security hardening, 348 tests\nLicense\n\nMIT — free for any AI agent, commercial or personal." }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/dmdorta1111/nima-core", "publisherUrl": "https://clawhub.ai/dmdorta1111/nima-core", "owner": "dmdorta1111", "version": "3.3.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/nima-core", "downloadUrl": "https://openagent3.xyz/downloads/nima-core", "agentUrl": "https://openagent3.xyz/skills/nima-core/agent", "manifestUrl": "https://openagent3.xyz/skills/nima-core/agent.json", "briefUrl": "https://openagent3.xyz/skills/nima-core/agent.md" } }