joanmarcriera/Home-office-automations
Agent Claude Code
Quality-check playbooks and tool docs against repo standards. Use when a user adds or edits a playbook or tool doc, or asks to review documentation quality. Checks structure, internal links, taxonomy placement, and standards compliance.
joanmarcriera/Home-office-automations
Hook Claude Code
Runs after a tool call finishes for Edit and Write tool calls, running an inline shell check. From joanmarcriera/Home-office-automations.
joanmarcriera/Home-office-automations
Hook Claude Code
Runs before the agent uses a tool for Edit and Write tool calls, running an inline shell check. From joanmarcriera/Home-office-automations.
joanmarcriera/Home-office-automations
Settings file Claude Code
Agent settings declaring 2 hook events (PostToolUse, PreToolUse).
joanmarcriera/Home-office-automations
Skill Claude CodeCodex
Ingest a new AI source into the knowledge base. Checks new-sources.md for unprocessed entries, classifies each by taxonomy, creates or updates the canonical tool doc, and marks the source as processed.
joanmarcriera/Home-office-automations
Skill Claude CodeCodex
Scaffold a new tool documentation page following the repo's catalogue conventions. Pass the tool name and category. Creates the file, populates the standard template, and adds it to mkdocs.yml nav.
joanmarcriera/Home-office-automations
Instructions file CodexOpenCode
AGENTS.md instructions for joanmarcriera/Home-office-automations, covering agents.md, objective, non-negotiable rules, section ownership map and standard task checklists.
joanmarcriera/Home-office-automations
Instructions file
Claude Code instructions for joanmarcriera/Home-office-automations, covering claude code — project memory, project purpose, primary maintenance docs, stack at a glance and key conventions (from docs/standards.md).
joanmarcriera/Home-office-automations
Agent
Agency-Agents is a comprehensive suite of 110+ specialized AI agent personas designed to transform generic coding assistants into a "complete AI agency." In early January 2027, it serves as a critical configuration and system prompting layer for next-generation IDE-based agents like Claude 5.1, GPT-5.5, Gemini 4.0…
joanmarcriera/Home-office-automations
Agent
Agency Swarm (v1.4+, early January 2027) is an open-source, multi-agent orchestration framework that simplifies the creation of collaborative agent teams organized like a professional company or department. While originally built on top of the OpenAI Assistants API, it has evolved into a robust, provider-agnostic…
joanmarcriera/Home-office-automations
Agent
The Agentic Automation Canvas (AAC) is a structured framework and open-source tool for the prospective design, governance, and evaluation of agentic AI systems. It provides a machine-readable "project contract" that bridges the gap between high-level user expectations and technical implementation. As of early January…
joanmarcriera/Home-office-automations
Agent
An Agentic Workbench is an integrated software pattern and operational environment designed to orchestrate human-in-the-loop (HITL) collaboration with autonomous multi-agent systems in early 2027. It provides unified, real-time control planes where human operators supervise, steer, and co-execute workflows alongside…
joanmarcriera/Home-office-automations
Agent
Agno (v3.x+, early January 2027) is an ultra-fast, lightweight Python framework designed for building production-grade multi-modal agents with persistent memory, semantic knowledge, and customizable tools. As the official successor to Phidata v2, Agno is engineered specifically for microsecond-overhead performance and…
joanmarcriera/Home-office-automations
Agent
Anthropic Agent Skills (v1.3+, early January 2027) are standardized, highly optimized, and encapsulated task recipes—consisting of specialized system prompts, executable scripts, and API tool declarations—that autonomous agents load dynamically to solve complex, domain-specific engineering problems. Originally…
joanmarcriera/Home-office-automations
Agent
AutoReason (v2027.1.x+, early January 2027) is an autonomous reasoning framework by Nous Research designed to enable LLMs to perform complex, multi-step logical tasks with minimal human intervention. It implements advanced "Reasoning-as-a-Service" patterns, allowing models like Nous Hermes 4 (Llama 4 based), Gemma 4…
joanmarcriera/Home-office-automations
Agent
AWS Dogwood is a robust policy management and safety-focused agentic framework designed to authorize, evaluate, and restrict LLM agent actions and tool calls. It provides an enterprise-grade mechanism to declare declarative security and access control boundaries for autonomous agents, ensuring that even if an agent…
joanmarcriera/Home-office-automations
Agent
The Bee Agent Framework (v1.6+, early January 2027) is an open-source framework by IBM Research for building, deploying, and orchestrating production-grade AI agents. It provides complete feature parity between TypeScript and Python, allowing for robust multi-agent systems with native Model Context Protocol (MCP 3.1)…
joanmarcriera/Home-office-automations
Agent
The Claude skills ecosystem is the growing collection of reusable skill packs, command libraries, FastMCP tool providers, and workflow repositories built around Claude Code, Anthropic's Agent SDK, and related coding-agent toolchains. It leverages Anthropic's native tool-calling capabilities and structured execution…
joanmarcriera/Home-office-automations
Agent
Cline (formerly Claude Dev) is an open-source, autonomous AI coding agent operating natively within VS Code, JetBrains, and terminal environments. It possesses broad access to the local filesystem, terminal execution, and an embedded browser instance for end-to-end web testing and visual verification. As of early…
joanmarcriera/Home-office-automations
Agent
Composio (v1.5+, early January 2027) is an enterprise-grade tool integration and authentication middleware platform that connects AI agents to over 250+ external SaaS applications, local utilities, and infrastructure services. Serving as a robust bridge between agentic frameworks and actual API execution endpoints, it…
joanmarcriera/Home-office-automations
Agent
DeerFlow (v2.2+, early January 2027) is an enterprise-grade open-source agentic deep-research workflow orchestrator developed by ByteDance. It is recognized as a premier reference architecture for building high-autonomy research and information-synthesis agents that leverage frontier models including Claude 5.6…
joanmarcriera/Home-office-automations
Agent
The Documentation Writer Skill is a specialized, developer-oriented automation tool built for AI coding agents (such as Claude Code, Cline, and Roo Code). Adhering to the universal, cross-agent SKILL.md specification, it automates the creation, continuous auditing, and maintenance of repository documentation. In early…
joanmarcriera/Home-office-automations
Agent
Gemini API Managed Agents is Google's fully managed agent orchestration platform, executing multi-turn reasoning, tool calling, background code execution, and native FastMCP 3.1 protocol integration inside isolated cloud sandboxes, utilizing Gemini 4.0 Ultra and Gemini 3.6 Flash.
joanmarcriera/Home-office-automations
Agent
Gemini Robotics (anchored by Gemini Robotics 3 and its embodied reasoning variant, Gemini Robotics ER 3) is Google DeepMind's flagship model family designed for physical hardware orchestration, humanoid whole-body control, and multi-robot fleet coordination. It operates as a Vision-Language-Action (VLA) foundation…