agent▌

Multi-agent architectures distribute work across multiple agent invocations, each with its own focused context. When designed well, this distribution enables capabilities beyond single-agent limits. When designed poorly, it introduces coordination overhead that negates benefits. The critical insight is that sub-agents exist primarily to isolate context, not to anthropomorphize role division.

Playwright-based browser automation with element refs and session persistence for AI agents. \n \n Provides 6 core functions: open (navigate + configure), snapshot (refresh element refs), interact (click/fill/drag/upload/scroll), screenshot, execute (JavaScript), and close \n Uses @e ref system for element targeting; refs invalidate after navigation and require re-snapshot to refresh \n Supports video recording with optional cursor indicator, proxy routing, file uploads, and drag-and-drop intera

When working on the Agents workbench layout, always follow these guidelines:

Manage multiple local CLI agents in parallel tmux sessions with task assignment and monitoring. \n \n Start, stop, and monitor agents running in isolated tmux sessions with log tailing and health checks \n Assign tasks to specific agents and track their execution output in real time \n Schedule recurring agent work via cron integration for automated workflows \n Requires tmux and Python 3; agents are configured in a local agents/ directory \n

Before dispatching, analyze the task systematically. Think through step by step:

Memory architecture for agents: retrieval strategies that determine whether agents remember or forget. \n \n Covers five memory types: short-term (context window), long-term (vector stores), working memory, episodic memory, and semantic memory, each suited to different information patterns \n Emphasizes retrieval as the core challenge; provides chunking strategies, embedding quality guidance, and metadata filtering to surface the right memories at decision time \n Includes anti-patterns like sto

Design AI agent tools with clear schemas, descriptions, and error handling that prevent hallucination and token waste. \n \n Focuses on JSON Schema best practices and description writing that helps LLMs understand tool intent and constraints, not just implementation details \n Covers tool validation, explicit error handling patterns, and recovery strategies that prevent silent failures and agent loops \n Includes guidance on the Model Context Protocol (MCP) standard for tool interoperability acr

Persistent, searchable memory bank for AI agents with automatic project documentation sync. \n \n Provides four core MCP tools: memory_search for querying by text/type/tags, memory_write for recording knowledge and decisions, memory_read for retrieving specific entries, and memory_stats for usage analytics \n Organizes memories by type (architecture, patterns, decisions) and supports custom tagging for flexible retrieval and organization \n Runs as an MCP server that syncs with your project work

Manage Twill Cloud Coding Agent workflows through the public v1 API. \n \n Create, list, retrieve, and manage tasks with support for custom branches, agent selection, and file attachments \n Stream job logs in real time via Server-Sent Events and cancel running jobs at any point \n Full lifecycle control: send follow-up messages, approve plans, archive tasks, and manage task state \n Schedule recurring coding tasks with cron expressions, timezone support, and pause/resume capabilities \n List re

Skill by ara.so — Daily 2026 Skills collection.