A CLI tool for managing tasks with persistent memory, designed for AI-assisted development workflows
Lazy, elegant terminal CLI for chatting with Claude / OpenAI / Gemini / Ollama, orchestrating multi-step LLM workflows, and running multi-agent Slack teams with cross-task memory. Banner-on-launch, slash-command ghost autocomplete, persistent sessions, lo
AI long-term task memory system - cross-session persistence, hierarchical tasks, decision records
High-priority task queue for Node.js and browsers
Increase memory limit for local node binaries ('max-old-space-size')
A fast, efficient Node.js Worker Thread Pool implementation
High-priority task queue for Node.js and browsers
In memory queue system prioritizing tasks
Lightweight core CLI surface for Claude Flow — memory + hooks commands only. Designed to load fast on cold npx cache (<5s) so plugin skills don't race the 30s MCP-startup timeout. The full @claude-flow/cli metapackage lazy-loads everything else on top of
micromark extension to support GFM task list items
mdast extension to parse and serialize GFM task list items
AI long-term task memory system - cross-session persistence, hierarchical tasks, decision records
Memory and task management MCP Server with Goal-Task-Memory architecture
A shim for the setImmediate efficient script yielding API
MongoDB Server for testing (core package, without autodownload). The server will allow you to connect your favourite ODM or client library to the MongoDB Server and run parallel integration tests isolated from each other.
MongoDB Server for testing (auto-download latest version). The server will allow you to connect your favourite ODM or client library to the MongoDB Server and run parallel integration tests isolated from each other.
Useful TypeScript utilities.
A simple tool to keep requests to be executed in order.
In-memory Node.js and browser job scheduler
Memory adapter for Better Auth
Memory adapter for catbox
task item extension for tiptap
task list extension for tiptap
Personal AI friend — single-daemon coding companion
pikuri-tasks gives a pikuri-core agent an in-memory task list it can use to plan and track multi-step work. A +Pikuri::Tasks::List+ holds the per-Agent state; four tools (+task_create+, +task_in_progress+, +task_completed+, +task_delete+) mutate it via content-as-identifier (no item IDs to hallucinate). +Pikuri::Tasks::Extension+ wires the list and tools onto an +Pikuri::Agent+ via +c.add_extension(...)+ inside the +Agent.new+ block. The list lives in process memory only — nothing is written to disk. Sub-agents do not inherit the parent's list (consistent with the "sub-agents do not inherit extensions" rule).
Provision and interact with AI agents on agentd.link via a simple Ruby API or CLI.
Ties durable_decorator to rails and provides a few helpful utility rake tasks that rely on the Rails ENV being loaded into memory
RakeServer is a library which allows Rake tasks to be run using client requests to a long-running rake server, which can eagerly load required code into memory for fast task invocation.
A plugin for delayed_job to schedule simple tasks. It does not need a seperate worker and reduce the memory usage.
Harmoni keeps configuration files in sync between memory and on disk for hot loading and other tasks.
Provides various utility tasks for PHP based sites that use capistrano for deployment. Currently provides a method to clear the APC to avoid encountering memory allocation errors.
Because Solr sometimes fails, it happens. It might be a maintenance work you have to do or just Out-Of-Memory problems. If you are running search-sensitive Rails app, you have to deal with it.This gem was developed to postpone your index tasks automatically into a sidekiq queue if Solr engine becomes unavailable
RCrewAI is a powerful Ruby framework for creating autonomous AI agent crews that collaborate to solve complex tasks. Build intelligent workflows with reasoning agents, tool usage, memory systems, and human oversight. Key Features: • Multi-Agent Orchestration: Create crews of specialized AI agents that work together • Multi-LLM Support: OpenAI GPT-4, Anthropic Claude, Google Gemini, Azure OpenAI, Ollama • Rich Tool Ecosystem: Web search, file operations, SQL databases, email, code execution, PDF processing • Agent Memory: Short-term and long-term memory for learning from past executions • Human-in-the-Loop: Interactive approval workflows and collaborative decision making • Advanced Task Management: Dependencies, retries, async execution, and context sharing • Hierarchical Teams: Manager agents that coordinate and delegate to specialist agents • Production Ready: Security controls, error handling, comprehensive logging, and monitoring • Ruby-First Design: Built specifically for Ruby developers with idiomatic patterns • CLI Tools: Command-line interface for creating and managing AI crews
Pampa is a Ruby library for async & distributing computing providing the following features: - cluster-management with dynamic reconfiguration (joining and leaving nodes); - distribution of the computation jobs to the (active) nodes; - error handling, job-retry and fault tolerance; - fast (non-direct) communication to ensure realtime capabilities. The Pampa framework may be widely used for: - large scale web scraping with what we call a "bot-farm"; - payments processing for large-scale ecommerce websites; - reports generation for high demanded SaaS platforms; - heavy mathematical model computing; and any other tasks that requires a virtually infinite amount of CPU computing and memory resources. Find documentation here: https://github.com/leandrosardi/pampa
pikuri-memory gives a pikuri-core agent durable, long-lived memory: facts about the user and their work that persist across conversations. It wires a +recall+ tool plus an automatic per-turn prefetch onto an agent via +c.add_extension Pikuri::Memory::Extension.new(...)+ inside the +Agent.new+ block — same opt-in shape as +pikuri-tasks+ / +pikuri-vectordb+. Recall is automatic and synchronous (embed + vector search, milliseconds); capture is automatic and asynchronous (an off-the-interaction-path extraction queue), so a turn never blocks on "what should I remember?". Storage is mem0 (https://github.com/mem0ai/mem0) reached over a thin Faraday HTTP client — the append-only +add+ / read-time +search+ model. Only the *user's own words* are fed to extraction (a write-side hygiene rule that structurally drops system/assistant/tool-sourced junk), and recalled context enters the chat as a +:system+ message so it is provenance-tagged and excluded from the next extraction pass. This release ships the Ruby client + extension + tool against a *bring-your-own* mem0 endpoint; a self-managed mem0 sidecar supervisor (the +ChromaServer+-style docker pattern) is a follow-on.
+pikuri+ is the convenience bundle for the pikuri AI-assistant toolkit. It ships no Ruby code of its own beyond a tiny entry file that +require+'s each sibling gem; +gem install pikuri+ pulls in pikuri-core, pikuri-extractors, pikuri-pdf, pikuri-skills, pikuri-tasks, pikuri-memory, pikuri-workspace, pikuri-code, pikuri-mcp, pikuri-subagents, pikuri-vectordb, and pikuri-assistant in one shot, and +require 'pikuri'+ boots all of them. Privacy-conscious users who want a minimal dependency tree to audit should install +pikuri-core+ directly and opt into the extension gems they actually need — same +bundle add+ pattern Rails users have always had. See each pikuri-* gem's README for its individual surface.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.