# Team Context Management (TCM) Team Context Management is the practice of curating, sharing, and maintaining AI context at the team level. It bridges [[Personal Context Management (PCM)]] (individual) and [[Enterprise Context Management (ECM)]] (organizational), providing the shared context that enables teams to work effectively with AI. All levels of context management use the same principles and techniques of [[Context Engineering]], just applied at different scales. TCM focuses on **members, processes, and priorities**: the team-specific context that shapes how people collaborate with AI. ## What TCM covers - **Team members and roles**: who is on the team, their responsibilities, and expertise - **Processes and workflows**: how the team works, reviews code, handles incidents, ships features - **Priorities**: what the team is focused on, current sprint goals, quarterly objectives - **Shared coding standards and conventions**: CLAUDE.md, AGENTS.md, and similar configuration files that encode team-specific rules, style guides, and architectural decisions - **Shared skills and workflows**: team-level [[AI Agent Skills|AI Skills]], templates, and procedures that standardize how AI assists across the team - **Team memory**: accumulated decisions, lessons learned, and patterns that persist across individual conversations - **Tool and MCP configurations**: shared tool setups, API integrations, and [[Model Context Protocol (MCP)|MCP]] server configurations ## Why TCM matters Without TCM, every team member manages their own AI context independently. This leads to: - Inconsistent AI outputs across the team - Duplicated effort in prompt engineering and context setup - Knowledge silos where one person's AI knows things another's doesn't - Difficulty onboarding new team members into AI-augmented workflows ## The context management hierarchy TCM sits in the middle of a nested hierarchy: - **[[Enterprise Context Management (ECM)|ECM]]**: organization-wide policies, compliance, standards (strategy, focus, culture) - **TCM**: team-level conventions and shared workflows (members, processes, priorities) - **[[Project Context Management (PCM)|Project Context Management]]**: project-specific context (architecture, design, implementation details, business rules) - **[[Personal Context Management (PCM)|PCM]]**: individual preferences, style, and personal knowledge [[Context Inheritance]] flows down: project context takes into account the enterprise context and team/service/department context. Teams inherit enterprise standards and add their own. Projects inherit team standards. Individuals layer their personal context on top. ## The form context takes The ideal form for team context is currently a set of Markdown files: - An **[[AI Master Prompt]]** that gives the ground rules and base context (e.g., team name, members, processes, goals, rules, priorities) - **AGENTS.md/CLAUDE.md files** at different levels of project source code, describing technical architecture, design, rules - **[[AI Agent Skills|AI Skills]]** describing how to perform certain tasks, workflows, and processes - **[[AI Agents]]** taking on specific roles and leveraging different combinations of skills ## Practical examples - A shared CLAUDE.md in a Git repository that encodes coding conventions, architecture decisions, and team practices - A team skill library that standardizes how AI handles code reviews, PR descriptions, or documentation - Shared MCP server configurations for team-specific tools and services - Team-level memory systems that capture decisions and their rationale ## References - ## Related - [[Personal Context Management (PCM)]] - [[Project Context Management (PCM)|Project Context Management]] - [[Enterprise Context Management (ECM)]] - [[Context Engineering]] - [[Context Inheritance]] - [[Context-as-Code]] - [[AI Master Prompt]] - [[How to create your Business AI Master Prompt]] - [[AI Agents]] - [[AI Agent Skills]] - [[Claude Code]] - [[Model Context Protocol (MCP)]] - [[Levels of AI Context Management]] - [[Harness Engineering]]