ai-dev-analytics 2.0.0 → 2.0.2

package/README.en.md

@@ -2,29 +2,35 @@
 
 # AIDA
 
-### Make project rules, skills, and module memory the real source of truth.
+### JSON truth-source management for AI tool assets.
 
-AIDA 2.0 is no longer centered on task ledgers or run timelines. It keeps the assets that actually matter over time:
-**rules, skills, module memory, and branch-level demand summaries.**
+AIDA 2.0 keeps only the assets that matter over time:
+**rules, skills, memories, and summaries.**
 
 ```json
 { "mcpServers": { "aida": { "command": "npx", "args": ["--registry=https://registry.npmjs.org/", "-y", "ai-dev-analytics", "mcp"] } } }
 ```
 
-[![npm version](https://img.shields.io/badge/npm-v2.0.0-0066ff)](https://www.npmjs.com/package/ai-dev-analytics)
-[![license](https://img.shields.io/github/license/LWTlong/ai-dev-analytics?color=%23333)](./LICENSE)
-[![node](https://img.shields.io/node/v/ai-dev-analytics?color=%23339933)](https://nodejs.org)
-[![tests](https://img.shields.io/badge/tests-passing-brightgreen)](#testing)
-
 [Chinese README](./README.md) · [Command Reference](./COMMANDS.md) · [Docs](./docs/INDEX.md)
 
 </div>
 
 ---
 
-## What AIDA 2.0 Keeps
+## What 2.0 Is
+
+AIDA 2.0 is not an analytics dashboard or a runtime ledger.
+
+It is a clean asset manager for:
+
+- project rules
+- project skills
+- module memories
+- demand summaries
+
+All durable data lives in `.aida/*.json`. Tool directories such as `.cursor`, `.claude`, `.codex`, and `.lingma` are generated projections.
 
-The durable truth sources are:
+## Truth Sources
 
 ```text
 .aida/
@@ -36,64 +42,44 @@ The durable truth sources are:
   memories/
     index.json
     modules/*.json
-  rules/*.md
 ```
 
-- `rules.json`: project-level technical rules
-- `skills.json`: project-level skills registry
-- `memories/index.json`: low-cost memory lookup index
-- `memories/modules/*.json`: module-level business memory
-- `summary.json`: branch / demand summaries
-
-## What AIDA 2.0 Discards
-
-2.0 intentionally removes the heavy 1.x runtime ledger model:
+2.0 intentionally drops the old 1.x runtime noise:
 
 - `run.json`
 - persistent task ledgers
-- timeline / events / workflow history
+- timeline / workflow / events
 - `.aida/runs/**`
 - `.aida/index.json`
 - `.aida/tool-configs.json`
 
-Migration is a cleaning rebuild, not a full carry-over.
-
-## Command Model
-
-The main 2.0 commands are:
+## Main Commands
 
 ```bash
 aida init
 aida sync
-aida build
 aida doctor
-aida migrate-legacy
+aida rules
+aida skills
+aida memory
+aida mcp
 ```
 
-- `init`: create the 2.0 truth-source layout
+- `init`: initialize 2.0 truth sources
 - `sync`: refresh memories, summaries, and generated tool artifacts
-- `build`: rebuild generated artifacts from JSON truth sources
 - `doctor`: inspect and normalize project state
-- `migrate-legacy`: migrate 1.x / legacy projects into the 2.0 structure
-
-## Context Recovery
 
-Memory recovery is intentionally layered:
+## Quick Start
 
-1. read `memories/index.json`
-2. read `summary.json`
-3. only read matched `memories/modules/*.json`
+Add AIDA to your project MCP config:
 
-This keeps token usage low and prevents loading the full memory corpus every time.
-
-## Notes
-
-- MCP is an optional structured execution channel, not the system foundation.
-- CLI and MCP should operate on the same JSON truth sources.
-- Built-in bundled workflow skills are no longer seeded by default in 2.0.
+```json
+{ "mcpServers": { "aida": { "command": "npx", "args": ["--registry=https://registry.npmjs.org/", "-y", "ai-dev-analytics", "mcp"] } } }
+```
 
-For full product explanation, migration behavior, and Chinese examples, use:
+Then:
 
-- [README.md](./README.md)
-- [COMMANDS.md](./COMMANDS.md)
-- [docs/AIDA-2.0-DESIGN.md](./docs/AIDA-2.0-DESIGN.md)
+```bash
+aida init
+aida sync
+```