From 0321f0ee910691a9c62004a6a67cecd759aebb2b Mon Sep 17 00:00:00 2001 From: omaiesh Date: Thu, 26 Feb 2026 15:22:56 +0100 Subject: [PATCH] Consolidate overview hero, remove Key Features section Co-Authored-By: Claude Sonnet 4.6 --- docs/web/overview.md | 30 +++++++----------------------- 1 file changed, 7 insertions(+), 23 deletions(-) diff --git a/docs/web/overview.md b/docs/web/overview.md index 3dfad8bc..427f50f1 100644 --- a/docs/web/overview.md +++ b/docs/web/overview.md @@ -6,18 +6,15 @@ permalink: /overview/

Overview

-

Rosetta provides shared prompts and rules that stay consistent across IDEs, coding agents, and models. It uses a classification-first and meta-prompting approach so teams can run project-specific workflows with predictable quality.

+

Rosetta is a control plane for AI coding agents that automates context setup, enforces consistent workflows, and manages engineering knowledge at the organization level — without sharing your source code. It solves the core problems teams face with AI-assisted development:

+
    +
  • Fragmented adoption and inconsistent execution across teams and tools
    • +
  • Missing business and technical context in day-to-day AI-assisted development
    • +
  • Weak governance, low visibility, and avoidable risk in AI-enabled SDLC
    • +
  • Reinvented workflows and slow onboarding between projects
    • +
-## What Rosetta Solves - -Modern AI coding agents require externalized, persistent context to maintain stable behavior across sessions and tasks. Creating and maintaining this context is typically manual, error-prone, and slow. Rosetta automates the initialization and ongoing maintenance of AI coding agent context for new and existing codebases — providing agent rules, skills, workflows, sub-agents, and instructions as explicit, versioned artifacts, managed centrally via MCP so teams across the organization can share and evolve agent context consistently. - -- Fragmented adoption and inconsistent execution across teams and tools. -- Missing business and technical context in day-to-day AI-assisted development. -- Weak governance, low visibility, and avoidable risk in AI-enabled SDLC. -- Reinvented workflows and slow onboarding between projects. - ## Benefits By Role
@@ -53,19 +50,6 @@ Modern AI coding agents require externalized, persistent context to maintain sta

Release Driven

Evolve safely with release-based governance and rollback-friendly instruction management.

-## Key Features - -
-

Unified Knowledge Hub

Business context, architecture, requirements, and rules are organized in one retrievable system.

-

RAGFlow Integration

Publishes instruction artifacts for semantic retrieval via MCP tools in coding sessions.

-

Smart Metadata and Incremental Updates

Uses tags and hash-based change detection to publish only modified files.

-

Built-in Guardrails

Includes approval gates, risk controls, and validation checkpoints for safer execution.

-

Reference SDLC

Provides a complete lifecycle by default with opt-out flexibility and room for controlled process experiments.

-

Adoption Visibility

Supports usage tracking by capability and helps identify promoters, blockers, and high-value rollout patterns.

-

Single-Command Onboarding

Supports fast initialization, upgrades, and project-level customization.

-

Community-Friendly

Open-source workflow with contribution paths for improvements to rules and guidance.

-
- ## Architecture Snapshot - **Content:** markdown instructions and project/business context.