project-map

Maps the yeaboi codebase modules, CLI flags, environment variables, and MCP server internals.

4|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/yeaboi-ai/yeaboi.ai --skill project-map-yeaboi-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-map
Source: https://github.com/yeaboi-ai/yeaboi.ai/tree/main/.claude/skills/project-map
Command: npx skills add https://github.com/yeaboi-ai/yeaboi.ai --skill project-map-yeaboi-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Navigating a large Python codebase like yeaboi (with its agent graph, MCP server, TUI, and many mode packages) is slow without a guide. This Skill provides an annotated map of every module in src/yeaboi/, the full CLI surface, and all environment variables so you can orient yourself instantly. ## Core Features & Use Cases - Annotated Module Map: Every file in src/yeaboi/ (agent, standup, retro, performance, agentwatch, reporting, mcp, tools, ui, repl) with a one-line description of its responsibility. - CLI & Config Reference: Complete table of CLI flags, subcommands, validation rules, and all environment variables with their defaults and effects. - MCP Server Internals: Documents the 26-tool MCP stdio server design, the sampling/provider/fallback LLM chain, and the Claude Code plugin layout. - Use Case: When adding a new CLI flag or modifying the MCP server, consult this map to find exactly which modules (cli.py, config.py, mcp/tools_*.py) need changes and which tests enforce surface parity. ## Quick Start Ask the AI to show you which modules handle standup delivery or where a specific environment variable is consumed in the yeaboi codebase.

Frequently Asked Questions about project-map

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I navigate the yeaboi codebase structure?

Use the project map's annotated listing of src/yeaboi/, which describes every module's responsibility in one line. Agent logic lives in agent/, prompts in prompts/, integrations in tools/, and the full-screen TUI in ui/.

Where do I add a new CLI flag or subcommand in yeaboi?

CLI flags and subcommands are defined in cli.py using argparse with add_subparsers. Subcommand handlers are thin _cmd_* functions in cli.py, with tests in tests/unit/test_cli_subcommands.py.

How does the yeaboi MCP server handle LLM calls?

The MCP server resolves an LLM mode chain: sampling (if the client supports it), then provider (if an API key is configured), then deterministic fallback. SamplingChatModel bridges MCP sampling to LangChain via a ContextVar override in agent/llm.py.

What environment variables configure yeaboi LLM providers?

Set LLM_PROVIDER to anthropic, openai, google, bedrock, ollama, or an OpenAI-wire vendor like xai or deepseek, plus the matching API key variable such as ANTHROPIC_API_KEY or OPENAI_API_KEY. LLM_MODEL optionally overrides the model.

Why must nothing print to stdout in the yeaboi MCP server?

The MCP server uses stdio transport, so stdout carries JSON-RPC messages exclusively. Any print to stdout would corrupt the protocol; logs go to ~/.yeaboi/logs/mcp/mcp.log instead, enforced by a capsys test.