{"id":48739629,"url":"https://github.com/terrizoaguimor/celiums-memory","last_synced_at":"2026-04-28T14:05:04.714Z","repository":{"id":350090552,"uuid":"1205249099","full_name":"terrizoaguimor/celiums-memory","owner":"terrizoaguimor","description":"Neuroscience-grounded persistent memory for AI — with emotions, circadian rhythm, and 5,100 expert knowledge modules. Try it: ask.celiums.ai","archived":false,"fork":false,"pushed_at":"2026-04-13T05:03:32.000Z","size":24077,"stargazers_count":3,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-04-13T07:28:34.096Z","etag":null,"topics":["ai","ai-memory","artificial-intelligence","circadian-rhythm","claude-code","cognitive-architecture","cursor","developer-tools","emotional-ai","knowledge-engine","mcp","memory","model-context-protocol","neuroscience","open-source","opencore","persistent-memory","personality","typescript","vscode"],"latest_commit_sha":null,"homepage":"https://ask.celiums.ai","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/terrizoaguimor.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-04-08T19:30:39.000Z","updated_at":"2026-04-13T05:03:39.000Z","dependencies_parsed_at":null,"dependency_job_id":"ebe7eac0-ae65-4760-8a31-de2261bb6a96","html_url":"https://github.com/terrizoaguimor/celiums-memory","commit_stats":null,"previous_names":["terrizoaguimor/celiums-memory"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/terrizoaguimor/celiums-memory","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/terrizoaguimor%2Fceliums-memory","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/terrizoaguimor%2Fceliums-memory/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/terrizoaguimor%2Fceliums-memory/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/terrizoaguimor%2Fceliums-memory/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/terrizoaguimor","download_url":"https://codeload.github.com/terrizoaguimor/celiums-memory/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/terrizoaguimor%2Fceliums-memory/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31787263,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-14T02:24:21.117Z","status":"ssl_error","status_checked_at":"2026-04-14T02:24:20.627Z","response_time":153,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","ai-memory","artificial-intelligence","circadian-rhythm","claude-code","cognitive-architecture","cursor","developer-tools","emotional-ai","knowledge-engine","mcp","memory","model-context-protocol","neuroscience","open-source","opencore","persistent-memory","personality","typescript","vscode"],"created_at":"2026-04-12T06:13:51.946Z","updated_at":"2026-04-28T14:05:04.708Z","avatar_url":"https://github.com/terrizoaguimor.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\n\u003cbr /\u003e\n\n[![celiums-memory MCP server](https://glama.ai/mcp/servers/terrizoaguimor/celiums-memory/badges/card.svg)](https://glama.ai/mcp/servers/terrizoaguimor/celiums-memory)\n\n# Celiums\n\n### Your AI doesn't know what it doesn't know. And it forgets everything.\n\n**The open-source engine that gives AI persistent memory and instant access to 5,100+ expert knowledge modules — with a biological clock that adapts to each user.**\n\n[Try the Live Demo](https://ask.celiums.ai) · [Quick Start](#-quick-start) · [6 Tools](#-the-6-tools) · [How to Use](#-how-to-use-it) · [Architecture](#-architecture) · [Deploy](#-deploy-modes) · [Docs](https://celiums.ai/docs)\n\n[![npm version](https://img.shields.io/npm/v/@celiums/memory?style=flat-square\u0026color=22c55e)](https://www.npmjs.com/package/@celiums/memory)\n[![Downloads](https://img.shields.io/npm/dw/@celiums/memory?style=flat-square\u0026color=22c55e)](https://www.npmjs.com/package/@celiums/memory)\n[![License](https://img.shields.io/github/license/terrizoaguimor/celiums-memory?style=flat-square)](LICENSE)\n[![TypeScript](https://img.shields.io/badge/TypeScript-007ACC?style=flat-square\u0026logo=typescript\u0026logoColor=white)](https://www.typescriptlang.org/)\n[![GitHub Stars](https://img.shields.io/github/stars/terrizoaguimor/celiums-memory?style=flat-square)](https://github.com/terrizoaguimor/celiums-memory)\n[![Glama](https://glama.ai/mcp/servers/terrizoaguimor/celiums-memory/badges/score.svg)](https://glama.ai/mcp/servers/terrizoaguimor/celiums-memory)\n\n\u003cbr /\u003e\n\n\u003c/div\u003e\n\n---\n\n\u003e **What's new in v1.2 — 2026-04-27**\n\u003e\n\u003e - 🆕 **20 MCP tools** (was 6): journal (5), write (7), research (8) added.\n\u003e - 🆕 **BYOK LLM** — bring your own OpenAI-compatible endpoint. Works with OpenAI, Ollama, OpenRouter, Together, Groq, vLLM, LM Studio. No proprietary lock-in.\n\u003e - 🆕 **Ethics Engine layers B + C** — CVaR-probabilistic risk scoring + 5-framework philosophical evaluation, on top of layer A.\n\u003e - 🆕 **Integration utilities** — encrypted credential storage (`integrations/crypto.ts`), schema for tenant integrations, opportunistic LLM-powered output formatting (`humanize.ts`), free-form-query intent classifier.\n\u003e - 🧹 OpenCore is **fully self-contained** — zero network calls if you don't configure an LLM. The engine boots clean with nothing but a database.\n\n---\n\n## The Problem\n\nEvery time your AI assistant starts a new session, it starts from zero. It doesn't remember your preferences, your project decisions, your debugging history, or what you were working on yesterday. It hallucinates because it has no specialized knowledge — just general training data frozen at a cutoff date.\n\n**You spend more time re-explaining context than getting work done.**\n\n## The Solution\n\nCeliums combines two engines into one:\n\n| Engine | What it does | How |\n|---|---|---|\n| **Memory** | Remembers everything — with emotion | PAD vectors, dopamine, circadian rhythm, 15 cognitive modules |\n| **Knowledge** | Knows what experts know | 5,100 curated technical modules, full-text search, 18 categories |\n\nBoth engines expose **6 MCP tools** that any AI IDE can call autonomously. Install once, your AI has persistent memory AND expert knowledge forever.\n\n### See it in action: [ask.celiums.ai](https://ask.celiums.ai)\n\n\u003e Talk to Celiums AI directly — it uses all 5,100 modules, remembers you across sessions, and has a real circadian rhythm. Zero-knowledge: your data is never used for training.\n\n---\n\n## Quick Start\n\n### Option 1: npm (local, 60 seconds)\n\n```bash\nnpm install -g @celiums/cli\nceliums init\n```\n\nThat's it. `celiums init`:\n- Asks your name, timezone, and if you're a morning or night person\n- Loads 5,100 expert knowledge modules\n- Auto-configures Claude Code, Cursor, and VS Code\n- Creates your personal cognitive profile (circadian rhythm adapts to YOU)\n\n### Option 2: Docker (VPS, 3 minutes)\n\n```bash\n# 1. Clone\ngit clone https://github.com/terrizoaguimor/celiums-memory.git\ncd celiums-memory\n\n# 2. Configure\ncp .env.example .env   # edit passwords\n\n# 3. Start infrastructure (PostgreSQL + Qdrant + Valkey)\ndocker compose up -d\n\n# 4. Install dependencies\npnpm install\n\n# 5. Build + start Celiums\npnpm setup\n```\n\nYou get: Celiums API on port 3210 + PostgreSQL + Qdrant + Valkey.\nOn first run, 5,100 expert modules are loaded automatically.\n\n### Option 3: DigitalOcean 1-Click (coming soon)\n\nOne button. Deploys everything on your own DO droplet.\n\n---\n\n## Configure your LLM (BYOK)\n\nOpenCore tools (recall, remember, forage, absorb, sense, map_network, synthesize, bloom, cultivate) work **without any LLM** — pure local memory + knowledge base.\n\nThe AI-backed tools (journal, write, research) require an **OpenAI-compatible** chat endpoint. You bring your own key. The engine never talks to a Celiums-hosted service for inference.\n\n```bash\n# Option A — OpenAI (default endpoint)\nexport CELIUMS_LLM_API_KEY=sk-...\n\n# Option B — Ollama (local, free, no API key)\nexport CELIUMS_LLM_BASE_URL=http://localhost:11434/v1\nexport CELIUMS_LLM_API_KEY=ollama\nexport CELIUMS_LLM_MODEL=llama3.2\n\n# Option C — OpenRouter (any model, one key)\nexport CELIUMS_LLM_BASE_URL=https://openrouter.ai/api/v1\nexport CELIUMS_LLM_API_KEY=sk-or-...\nexport CELIUMS_LLM_MODEL=anthropic/claude-3.5-sonnet\n\n# Option D — Together / Groq / Anyscale / vLLM / LM Studio\n# Same pattern: set BASE_URL + API_KEY + (optional) MODEL.\n```\n\n| Env var | Default | Purpose |\n|---|---|---|\n| `CELIUMS_LLM_BASE_URL` | `https://api.openai.com/v1` | OpenAI-compatible endpoint root |\n| `CELIUMS_LLM_API_KEY` | *(empty — required to enable AI tools)* | Bearer token for the endpoint |\n| `CELIUMS_LLM_MODEL` | `gpt-4o-mini` | Default chat model |\n| `CELIUMS_EMBED_MODEL` | `text-embedding-3-small` | Default embedding model |\n| `CELIUMS_SEARCH_URL` | *(empty — optional)* | Corpus-search backend for `research_*` |\n\nIf `CELIUMS_LLM_API_KEY` is not set, AI-backed tools are simply not registered — `tools/list` returns only OpenCore. The engine never errors at boot for missing optional config.\n\n---\n\n## The Tools\n\nWhen connected via MCP, your AI can call these autonomously. Tools split into **OpenCore** (always available, no LLM required) and **AI-backed** (require an OpenAI-compatible LLM key — see *Configure your LLM* below).\n\n### Knowledge tools — OpenCore\n\n| Tool | What it does | Example |\n|---|---|---|\n| `forage` | Search for expert knowledge | *\"find modules about Kubernetes security\"* |\n| `absorb` | Load a specific module | *\"load the react-server-components module\"* |\n| `sense` | Get recommendations for a goal | *\"what should I use for building a REST API?\"* |\n| `map_network` | Browse all categories | *\"show me what knowledge areas are covered\"* |\n\n### Memory tools — OpenCore\n\n| Tool | What it does | Example |\n|---|---|---|\n| `remember` | Store something in memory | *\"remember that we chose Hono over Express\"* |\n| `recall` | Retrieve by semantic relevance | *\"what framework decisions did we make?\"* |\n| `synthesize` | Consolidate memories into a narrative | *\"what did I learn this week?\"* |\n| `bloom` | Expand a concept into related ideas | *\"explore variations of memory consolidation\"* |\n| `cultivate` | Deep-dive a topic | *\"cultivate hybrid retrieval\"* |\n\n### Journal tools — *AI-backed* (since v1.2)\n\nPersistent agent diary that survives across discontinuous invocations. Every model carries its own journal — when a new model takes over, it can *read* the predecessor's entries but never claim it lived them.\n\n| Tool | What it does |\n|---|---|\n| `journal_write` | Append a new entry (auto-embedded, importance-scored) |\n| `journal_recall` | Semantic + tag + type search across the agent's history |\n| `journal_arc` | Build a coherent arc with anti-confabulation guardrails |\n| `journal_introspect` | Answer a self-question grounded in entries only |\n| `journal_dialogue` | The agent reacts to a user-shared entry |\n\n### Write tools — *AI-backed* (since v1.2)\n\nNovelist-grade project state. Tracks `secrets_known_at_chapter` per character, worldbuilding rules with cost/exceptions, and timeline markers — flags structural continuity issues, not line-by-line prose problems.\n\n`write_project_create`, `write_project_get`, `write_character_create`, `write_scene_create`, `write_scene_update`, `write_continuity_check`, `write_export`.\n\n### Research tools — *AI-backed* (since v1.2)\n\nPersistent multi-session investigations with citations, findings, and gaps. Resume a project days later and see all prior context in one shot.\n\n`research_project_create`, `research_project_list`, `research_project_continue`, `research_finding_add`, `research_gap_add`, `research_search`, `research_synthesize`, `research_export`.\n\n\u003e `research_search` and `research_synthesize` need a corpus-search backend (`CELIUMS_SEARCH_URL`, any service exposing `POST /v1/search`). Without it the project/findings/gaps trackers still work fine.\n\n**What happens behind `remember`** (the user sees nothing, it just works):\n\n```\nUser: \"remember that we chose Hono over Express for the API\"\n                    |\n          PAD Emotional Vector (pleasure: 0.4, arousal: 0.3, dominance: 0.5)\n                    |\n          Theory of Mind (empathy matrix transforms user emotion)\n                    |\n          Dopamine / Habituation (novelty detection, reward modulation)\n                    |\n          Per-User Circadian (your timezone, your peak hour, your rhythm)\n                    |\n          PFC Regulation (clamp safe bounds, suppress extremes)\n                    |\n          Triple-Store Persist (PostgreSQL + Qdrant + Valkey)\n                    |\n          \"Remembered (importance: 0.72)\"\n```\n\n15 cognitive systems fire on a single `remember` call. The user just types one sentence.\n\n---\n\n## How to Use It\n\n### Connect to your IDE\n\nAfter `celiums init`, it's auto-wired. Or manually:\n\n**Claude Code:**\n```bash\nclaude mcp add celiums -- celiums start --mcp\n```\n\n**Cursor** — add to `~/.cursor/mcp.json`:\n```json\n{\n  \"mcpServers\": {\n    \"celiums\": { \"command\": \"celiums\", \"args\": [\"start\", \"--mcp\"] }\n  }\n}\n```\n\n**VS Code** — add to settings.json:\n```json\n{\n  \"mcp.servers\": {\n    \"celiums\": { \"type\": \"stdio\", \"command\": \"celiums\", \"args\": [\"start\", \"--mcp\"] }\n  }\n}\n```\n\n### Use the tools in conversation\n\nOnce connected, your AI uses the tools automatically. Just talk normally:\n\n```\nYou: \"Find me best practices for PostgreSQL optimization\"\nAI:  -\u003e calls forage(query=\"PostgreSQL optimization\")\n     -\u003e finds postgresql-best-practices-v2 (eval: 4.0)\n     -\u003e presents the expert module content\n\nYou: \"Remember that we decided to use JSONB for metadata columns\"\nAI:  -\u003e calls remember(content=\"decided to use JSONB for metadata columns\")\n     -\u003e stored with importance 0.68, mood: focused\n\nYou: \"What database decisions have we made?\"\nAI:  -\u003e calls recall(query=\"database decisions\")\n     -\u003e finds: \"decided to use JSONB for metadata\" (score: 0.89)\n     -\u003e presents with emotional context\n```\n\n### REST API\n\nIf running as a server (Docker/VPS), the full API is available:\n\n```bash\n# Search modules\ncurl http://localhost:3210/v1/modules?q=react+hooks\n\n# Get a specific module\ncurl http://localhost:3210/v1/modules/typescript-mastery\n\n# Browse categories\ncurl http://localhost:3210/v1/categories\n\n# Store a memory\ncurl -X POST http://localhost:3210/store \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"content\": \"The API uses Hono framework\", \"userId\": \"dev1\"}'\n\n# Recall memories\ncurl -X POST http://localhost:3210/recall \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"query\": \"what framework\", \"userId\": \"dev1\"}'\n\n# Check your circadian rhythm\ncurl http://localhost:3210/circadian?userId=dev1\n\n# Update your timezone\ncurl -X PUT http://localhost:3210/profile \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"userId\": \"dev1\", \"timezoneIana\": \"Asia/Tokyo\", \"timezoneOffset\": 9}'\n\n# MCP protocol (for AI clients)\ncurl -X POST http://localhost:3210/mcp \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/list\"}'\n\n# Health check\ncurl http://localhost:3210/health\n```\n\n### Configuration\n\nAll settings via environment variables:\n\n```bash\n# Core\nDATABASE_URL=postgresql://user:pass@localhost:5432/celiums_memory\nQDRANT_URL=http://localhost:6333\nVALKEY_URL=redis://localhost:6379\nPORT=3210\n\n# SQLite mode (alternative, single file, zero infrastructure)\nSQLITE_PATH=./celiums.db\n\n# Knowledge engine\nKNOWLEDGE_DATABASE_URL=postgresql://user:pass@localhost:5432/celiums\n\n# Onboarding (auto-configure on first run)\nCELIUMS_USER_NAME=dev1\nCELIUMS_LANGUAGE=en     # en, es, pt-BR, zh-CN, ja\nCELIUMS_TIMEZONE=America/New_York\nCELIUMS_CHRONOTYPE=morning  # morning, neutral, night\n```\n\n---\n\n## Architecture\n\n```\nYour AI (Claude Code, Cursor, VS Code, any MCP client)\n         |\n         | MCP JSON-RPC (6 tools)\n         v\n  CELIUMS ENGINE (1 process, 1 port)\n  |                              |\n  |  Knowledge Engine            |  Memory Engine\n  |  forage, absorb,             |  remember, recall\n  |  sense, map_network          |\n  |                              |  15 cognitive modules:\n  |  5,100 modules               |  limbic, circadian, dopamine,\n  |  18 dev categories           |  personality, ToM, PFC, ANS,\n  |  full-text search            |  habituation, reward,\n  |                              |  interoception, consolidation,\n  |                              |  lifecycle, autonomy,\n  |                              |  recall engine, importance\n  |                              |\n  v                              v\n  Modules DB                     Memory DB\n  (SQLite or PostgreSQL)         (SQLite or PG + Qdrant + Valkey)\n```\n\n### Per-User Circadian Rhythm\n\nEach user gets their own biological clock:\n\n```bash\ncurl http://localhost:3210/circadian?userId=dev1\n# {\n#   \"localHour\": 10.5,\n#   \"rhythmComponent\": 0.99,\n#   \"timeOfDay\": \"morning-peak\",\n#   \"circadianContribution\": 0.30\n# }\n```\n\nA user in Tokyo gets different arousal than a user in New York at the same moment.\n\n### Capability Gating\n\nTools appear based on your configuration. No upgrade prompts, no locked features visible.\n\n| Tier | Tools | What you get |\n|---|---|---|\n| **OpenCore** (free) | 6 | forage, absorb, sense, map_network, remember, recall + 5,100 modules |\n| **+ Fleet** (coming) | +8 | synthesize, bloom, cultivate, pollinate, decompose, fleet, construct |\n| **+ Atlas** (coming) | +12 | Real-time collaboration, 451K+ modules |\n\n---\n\n## Deploy Modes\n\n### Local (SQLite)\n\n```bash\nSQLITE_PATH=./celiums.db celiums start\n```\n\nEverything in one file. Perfect for individual developers.\n\n### Docker (full stack)\n\n```bash\ndocker compose up -d\n```\n\nPostgreSQL 17 + pgvector, Qdrant, Valkey. Optional Cloudflare Tunnel:\n\n```bash\ndocker compose --profile tunnel up -d\n```\n\n### DigitalOcean 1-Click (coming soon)\n\nOne button creates a droplet with everything pre-configured.\n\n---\n\n## Languages\n\n| | Language | Status |\n|---|---|---|\n| English | Default |\n| Espanol | Supported |\n| Portugues (Brasil) | Supported |\n| Chinese (Simplified) | Supported |\n| Japanese | Supported |\n\nAuto-detected from your OS during `celiums init`.\n\n---\n\n## Packages\n\n| Package | Description |\n|---|---|\n| `@celiums/memory` | Cognitive engine (15 modules, PAD, circadian) |\n| `@celiums/memory-types` | TypeScript types |\n| `@celiums/modules-starter` | 5,100 curated expert modules |\n| `@celiums/core` | Knowledge engine (search, modules, tools) |\n| `@celiums/cli` | CLI (`celiums init`, `celiums start`) |\n| `@celiums/adapter-mcp` | MCP protocol adapter |\n| `@celiums/adapter-rest` | REST API adapter |\n| `@celiums/adapter-openai` | OpenAI Function Calling adapter |\n| `@celiums/adapter-a2a` | Google A2A protocol adapter |\n\n---\n\n## Security\n\n- **Local-first.** Your memories live ONLY on your machine or your own server. Nothing is sent to us.\n- **API key auth.** Bearer token required for all non-localhost requests.\n- **Per-user isolation.** Each user has their own memory space, emotional state, and circadian profile.\n- **No telemetry.** Zero analytics, zero tracking, zero phone-home.\n\n---\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md).\n\n```bash\ngit clone https://github.com/terrizoaguimor/celiums-memory.git\ncd celiums-memory\npnpm install\npnpm build\n```\n\n---\n\n## Support This Project\n\nThis project is built on ADHD hyperfocus, too much coffee, and the stubborn belief that AI deserves a real brain. Every one of these 11,000+ lines was written between 20-hour coding sessions, fueled by curiosity and obsession.\n\nIf Celiums is useful to you, or if you believe AI should have emotions and not just compute, consider supporting the work.\n\n\u003ca href=\"https://celiums.ai/support\"\u003e\n  \u003cimg src=\"https://img.shields.io/badge/Support%20Celiums-Donate-green?style=for-the-badge\u0026logo=stripe\u0026logoColor=white\" alt=\"Support Celiums\" /\u003e\n\u003c/a\u003e\n\nYour contribution keeps the GPUs running, the coffee flowing, and this project alive.\n\n---\n\n## License\n\nApache 2.0 — see [LICENSE](LICENSE)\n\n---\n\n\u003cdiv align=\"center\"\u003e\n\n**Built with obsessive attention to detail.**\n\n[celiums.ai](https://celiums.ai) · [npm](https://www.npmjs.com/package/@celiums/memory) · [GitHub](https://github.com/terrizoaguimor/celiums-memory)\n\n\u003c/div\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fterrizoaguimor%2Fceliums-memory","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fterrizoaguimor%2Fceliums-memory","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fterrizoaguimor%2Fceliums-memory/lists"}