rezee

Gives AI agents access to a persistent document workspace so they can search, read, and update knowledge through 22 focused tools, 2 prompts, and 3 resources, with batch actions and direct workspace-item inspection. Runs either as a hosted Streamable HTTP endpoint with OAuth or API keys, or as a self-hosted stdio bridge pointed at your own instance.

Enables MCP-compatible AI development tools to discover and call workspace management tools over Git-friendly Markdown/YAML data, including project/task/knowledge/decision/evidence context and full-text search. It also exposes read-only and planning tools for legacy Buildroot workspaces that inspect status, history, sources, branches, worktrees, agents, and generate build/test/deploy/release plans without executing them or modifying the workspace.

Gives AI agents access to a persistent document workspace, letting them search, read, and update notes, guides, and project documentation across a full catalog of tools, prompts, and resources. It connects either to the hosted Cloud endpoint over Streamable HTTP with OAuth or API-key auth, or to a self-hosted instance through a stdio bridge.

Enables users to list workspaces and inspect concise pull-request review findings and hosted run summaries, and to prepare then, after explicit confirmation, start or rerun an existing hosted review. It deliberately never exposes raw diffs, full reports, artifacts, credentials, or prompt text.