organization-process

A product's specification and its playbook: Atlas, who carries a product from an idea to a published knowledge base, the PRD captured section by section through the curate gates, and the specification page of the product's studio, read from the graph.

3 skills

The agent pattern for the parts of a product that people give work to through an agent rather than a screen: the jobs they hand it, its tools and what each changes, what it reads, its guardrails and who confirms, how it talks, its hand-offs and the test conversations that prove each job, traced to the requirements and components of the architecture.

2 skills

Domain-driven design for the parts of a product that follow it: subdomains, bounded contexts and their map, event storming (events, commands, reactions, read models, external systems) and the tactical model (aggregates, entities, value objects, business rules, domain services), traced to the requirements and components of the architecture.

2 skills

Where a product's parts run and what they are built with: the cloud and its accounts or projects, each environment's region, console, logs and infrastructure code, the service under each part, the stack with versions, secrets by where they are kept, the pipelines, and who decides each choice (a named person, or the builder within limits). Its briefs stop an implementing agent on any missing detail.

2 skills

A product's architecture, pattern-neutral: its context and drivers, decisions with their alternatives, systems, components, interfaces, data stores and environments, each traced to the requirement it satisfies; each part handed to the design-pattern pack it follows, found among the project's packs.

2 skills

What is planned, by whom, blocked by what: work items and milestones over the product's requirements, features and components; what slips when a component is late, what is overdue, what waits on a decision.

2 skills

AGENTS.md as project instructions, by one option. Under claude-md-or-agents-md a project with no instruction files of its own gets its AGENTS.md files instead: a prompt.context hook asks $.fs.ancestors for every AGENTS.md and .claude/AGENTS.md from the filesystem root down to the working directory and hands them to the engine as project instruction files, which it renders, announces and withholds from agents exactly as it does CLAUDE.md, skipping any file the engine already loaded by import or link; a tool.call hook on Read attaches the ones under the project root as the engine attaches a nested CLAUDE.md. Under claude-md-and-agents-md every AGENTS.md is loaded beside CLAUDE.md. Under managed-only the project's and the person's instruction files are dropped and the organization's managed ones kept. Under claude-md the engine's walk stands alone. claude-md-or-agents-md is the default.

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

AI-powered codebase understanding — analyze, visualize, and explain any project

9 skills · 10 agents · hooks

Knowledge graph construction — entity extraction, relation mapping, and pathfinder graph traversal

2 skills · 1 command · 1 agent

Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.

12 skills · 1 MCP server · hooks

Work with your Box content directly from Claude Code — search files, organize folders, collaborate with your team, and use Box AI to answer questions, summarize documents, and extract data without leaving your workflow.

9 skills