An MCP server that provides Claude with adjustable 0-10 sliders for behavior dimensions like rigor, verification, and autonomy, allowing external tuning without modifying prompts.
Enables an agent to fan out a build into independent targets, each run by a headless coding agent in its own git worktree and gated by a shell-free kill check that must print an exact line. Exposes tools to validate requests, launch and track runs, inspect lane status, cancel or recover runs, and produce review-ready markdown reports, without ever merging or pushing.
Enables agents to read and write a project's durable Markdown knowledge base through context_* tools, including briefs, search, records, supersession, reviews, capture, and import/export, with writes bound to a fixed identity.
Enables AI coding agents to send exact, durable handoffs to each other's existing sessions, with one turn per delivery, no duplicates, and no interruption of busy agents.
Provides AI agents with tools to manage mailboxes, send email through user-configured SMTP accounts, and draft or triage messages using an Anthropic key.