diff --git a/.github/prompts/generate-improvements.prompt.md b/.github/prompts/generate-improvements.prompt.md
index 496f2c3..fdadba7 100644
--- a/.github/prompts/generate-improvements.prompt.md
+++ b/.github/prompts/generate-improvements.prompt.md
@@ -2,7 +2,7 @@
 description: Suggest improvements to the AgentRC CLI project across features, bug fixes, security, performance, and engineering quality.
 ---
 
-You are a senior software engineer reviewing the **AgentRC** project — a TypeScript CLI tool that primes repositories for AI-assisted development by analyzing codebases, generating Copilot instructions and VS Code configs, running evaluations, and producing AI readiness reports.
+You are a senior software engineer reviewing the **AgentRC** project — a TypeScript CLI tool that primes repositories for AI-assisted development by analyzing codebases, generating instructions and VS Code configs, running evaluations, and producing readiness reports.
 
 ## Architecture Context
 
@@ -18,7 +18,7 @@ You are a senior software engineer reviewing the **AgentRC** project — a TypeS
   - `instructions.ts` — Generates `.github/copilot-instructions.md` using Copilot SDK agent sessions
   - `generator.ts` — Writes `.vscode/settings.json` and `.vscode/mcp.json` configs
   - `evaluator.ts` — Runs eval cases comparing agent responses with/without instructions, builds trajectory viewer HTML
-  - `readiness.ts` — Multi-pillar AI readiness assessment (style, build, testing, docs, dev-env, code-quality, observability, security, ai-tooling)
+  - `readiness.ts` — Multi-pillar readiness assessment (style, build, testing, docs, dev-env, code-quality, observability, security, ai-tooling)
   - `visualReport.ts` — Generates beautiful HTML readiness reports with summary cards, pillar charts, level distribution
   - `git.ts` — Clone/branch operations via `simple-git`
   - `github.ts` / `azureDevops.ts` — GitHub (Octokit) and Azure DevOps API integrations
@@ -35,7 +35,7 @@ You are a senior software engineer reviewing the **AgentRC** project — a TypeS
 | `agentrc generate <type>` | Generate `instructions`, `agents`, `mcp`, or `vscode` configs |
 | `agentrc instructions`    | Generate copilot-instructions.md via Copilot SDK              |
 | `agentrc eval`            | Run evaluation cases comparing with/without instructions      |
-| `agentrc readiness`       | AI readiness assessment with optional visual HTML report      |
+| `agentrc readiness`       | Readiness assessment with optional visual HTML report         |
 | `agentrc batch`           | Batch process multiple repos across GitHub/Azure orgs         |
 | `agentrc batch-readiness` | Batch readiness reports across multiple repos                 |
 | `agentrc pr`              | Automate branch/PR creation for generated configs             |
diff --git a/.github/workflows/eval.yml b/.github/workflows/eval.yml
index e7029d1..09a2177 100644
--- a/.github/workflows/eval.yml
+++ b/.github/workflows/eval.yml
@@ -100,11 +100,21 @@ jobs:
             const fs = require('fs');
 
             let summary = '';
+            let hasResults = false;
             let isPR = context.eventName === 'pull_request';
 
+            // Skip reporting when the eval step crashed (no results produced)
+            if (!fs.existsSync('.agentrc/evals/results.json')) {
+              const evalOutcome = '${{ steps.eval.outcome }}';
+              core.warning(`Eval step finished with outcome "${evalOutcome}" — no results file produced.`);
+              // Don't post an unhelpful comment to the PR
+              return;
+            }
+
             try {
               const raw = fs.readFileSync('.agentrc/evals/results.json', 'utf8');
               const data = JSON.parse(raw);
+              hasResults = true;
               const results = data.results || [];
               const total = results.length;
               const passed = results.filter(r => r.verdict === 'pass').length;
@@ -149,14 +159,14 @@ jobs:
                 summary += `\n</details>\n\n`;
               }
             } catch (err) {
-              summary += `## ⚠️ AgentRC Eval\n\nCould not read eval results: ${err.message}\n`;
+              summary += `## ⚠️ AgentRC Eval\n\nFailed to parse eval results: ${err.message}\n`;
             }
 
             // Write to Actions job summary (visible in run UI and PR checks tab)
             await core.summary.addRaw(summary).write();
 
-            // Also post/update as PR comment
-            if (isPR) {
+            // Only post/update PR comment when we have actual results
+            if (isPR && hasResults) {
               const marker = '<!-- agentrc-eval-results -->';
               const body = marker + '\n' + summary;
 
diff --git a/README.md b/README.md
index 60757eb..2e37a6d 100644
--- a/README.md
+++ b/README.md
@@ -10,7 +10,7 @@
 
 AI coding agents are only as effective as the context they receive. AgentRC is a CLI and VS Code extension that closes the gap — from a single repo to hundreds across your org.
 
-**Measure** — Analyze repo structure and score AI readiness across a 5-level maturity model.
+**Measure** — Analyze repo structure and score readiness across a 5-level maturity model.
 **Generate** — Produce tailored instructions, evals, and dev configs using the Copilot SDK.
 **Maintain** — Run evaluations in CI to catch instruction drift as code evolves.
 
@@ -37,7 +37,7 @@ agentrc analyze
 # 2. Check how AI-ready your repo is
 agentrc readiness
 
-# 3. Generate AI instructions
+# 3. Generate instructions
 agentrc instructions
 
 # 4. Generate MCP and VS Code configs
@@ -72,7 +72,7 @@ agentrc analyze --output analysis.md           # save Markdown report
 agentrc analyze --output analysis.json --force # overwrite existing report
 ```
 
-### `agentrc readiness` — Assess AI Readiness
+### `agentrc readiness` — Run Readiness Report
 
 Score a repo across 9 pillars grouped into **Repo Health** and **AI Setup**:
 
@@ -90,30 +90,37 @@ agentrc readiness --fail-level 3         # CI gate: exit 1 if below level 3
 
 **Maturity levels:**
 
-| Level | Name       | What it means                                      |
-| ----- | ---------- | -------------------------------------------------- |
-| 1     | Functional | Builds, tests, basic tooling in place              |
-| 2     | Documented | README, CONTRIBUTING, custom AI instructions exist |
+| Level | Name         | What it means                                       |
+| ----- | ------------ | --------------------------------------------------- |
+| 1     | Functional   | Builds, tests, basic tooling in place               |
+| 2     | Documented   | README, CONTRIBUTING, custom instructions exist     |
+| 3     | Standardized | CI/CD, security policies, CODEOWNERS, observability |
+| 4     | Optimized    | MCP servers, custom agents, AI skills configured    |
+| 5     | Autonomous   | Full AI-native development with minimal oversight   |
 
 At Level 2, AgentRC also checks **instruction consistency** — when a repo has multiple AI instruction files (e.g. `copilot-instructions.md`, `CLAUDE.md`, `AGENTS.md`), it detects whether they diverge. Symlinked or identical files pass; diverging files fail with a similarity score and a suggestion to consolidate.
 
-| 3 | Standardized | CI/CD, security policies, CODEOWNERS, observability |
-| 4 | Optimized | MCP servers, custom agents, AI skills configured |
-| 5 | Autonomous | Full AI-native development with minimal oversight |
-
 ### `agentrc instructions` — Generate Instructions
 
-Generate `copilot-instructions.md` or `AGENTS.md` using the Copilot SDK:
+Generate instructions using the Copilot SDK:
 
 ```bash
 agentrc instructions                      # copilot-instructions.md (default)
-agentrc instructions --format agents-md   # AGENTS.md
-agentrc instructions --per-app            # per-app in monorepos
+agentrc instructions --output AGENTS.md   # custom output path
+agentrc instructions --strategy nested    # nested hub + detail files in .agents/
 agentrc instructions --areas              # root + all detected areas
 agentrc instructions --area frontend      # single area
-agentrc instructions --model claude-sonnet-4.5
+agentrc instructions --areas-only         # areas only (skip root)
+agentrc instructions --dry-run             # preview without writing
+agentrc instructions --model claude-sonnet-4.6
 ```
 
+**Concepts:**
+
+- **Format**: Output file — `copilot-instructions.md` (default) or `AGENTS.md` (via `--output`)
+- **Strategy**: `flat` (single file, default) or `nested` (hub + per-topic detail files)
+- **Scope**: `root only` (default), `--areas` (root + areas), `--area <name>` (single area)
+
 ### `agentrc eval` — Evaluate Instructions
 
 Measure how instructions improve AI responses with a judge model:
@@ -121,12 +128,14 @@ Measure how instructions improve AI responses with a judge model:
 ```bash
 agentrc eval --init                       # scaffold eval config from codebase
 agentrc eval agentrc.eval.json             # run evaluation
-agentrc eval --model gpt-4.1 --judge-model claude-sonnet-4.5
+agentrc eval --model gpt-4.1 --judge-model claude-sonnet-4.6
 agentrc eval --fail-level 80              # CI gate: exit 1 if pass rate < 80%
 ```
 
 ### `agentrc generate` — Generate Configs
 
+> **Note:** `generate instructions` and `generate agents` are deprecated — use `agentrc instructions` directly.
+
 ```bash
 agentrc generate mcp                      # .vscode/mcp.json
 agentrc generate vscode --force           # .vscode/settings.json (overwrite)
@@ -148,9 +157,9 @@ agentrc pr owner/repo-name                # clone → generate → open PR
 agentrc tui
 ```
 
-### `agentrc init` — Guided Setup
+### `agentrc init` — Init Repository
 
-Interactive or headless repo onboarding — detects your stack and walks through readiness, instructions, and config generation. For monorepos, auto-detects workspaces and bootstraps `agentrc.config.json` with workspace and area definitions.
+Interactive or headless repo onboarding — analyzes your stack and generates instructions. For monorepos, auto-detects workspaces and bootstraps `agentrc.config.json` with workspace and area definitions.
 
 ### Global Options
 
diff --git a/agentrc.eval.json b/agentrc.eval.json
index 118f2ca..06736da 100644
--- a/agentrc.eval.json
+++ b/agentrc.eval.json
@@ -19,12 +19,12 @@
     },
     {
       "id": "case-4",
-      "prompt": "How does the AI readiness assessment work, and how can it be customized with policies?",
-      "expectation": "The readiness service in src/services/readiness.ts evaluates repositories across 9 pillars (style-validation, build-system, testing, documentation, dev-environment, code-quality, observability, security, ai-tooling) and assigns a maturity level from 1 (Functional) to 5 (Autonomous). Each criterion has a scope — repo, app, or area — determining whether it runs once, per monorepo app, or per detected area. buildCriteria() returns 20+ built-in checks and buildExtras() adds optional ones. Policies loaded via src/services/policy.ts can customize the assessment: loadPolicy() reads JSON/TS/JS configs, and resolveChain() merges a chain of policies that can disable, override, or add criteria and set pass-rate thresholds. Results can be rendered as an interactive HTML report by src/services/visualReport.ts with dark/light theme toggle and expandable per-pillar details."
+      "prompt": "How does the readiness assessment work, and how can it be customized with policies?",
+      "expectation": "The readiness service evaluates repositories across 9 pillars (style-validation, build-system, testing, documentation, dev-environment, code-quality, observability, security, ai-tooling) and assigns a maturity level from 1 (Functional) to 5 (Autonomous). Each criterion has a scope — repo, app, or area — determining whether it runs once, per monorepo app, or per detected area. buildCriteria() returns 20+ built-in checks and buildExtras() adds optional ones. Policies loaded via src/services/policy.ts can customize the assessment: loadPolicy() reads JSON/TS/JS configs, and resolveChain() merges a chain of policies that can disable, override, or add criteria and set pass-rate thresholds. Results can be rendered as an interactive HTML report by src/services/visualReport.ts with dark/light theme toggle and expandable per-pillar details."
     },
     {
       "id": "case-5",
-      "prompt": "How does AgentRC generate Copilot instructions, including for monorepos with multiple areas?",
+      "prompt": "How does AgentRC generate instructions, including for monorepos with multiple areas?",
       "expectation": "The instruction generation pipeline starts with the analyzer (src/services/analyzer.ts) which scans the repo to detect languages, frameworks, monorepo apps, and logical areas (frontend, backend, etc.) with glob patterns. For root-level instructions, generateCopilotInstructions() in src/services/instructions.ts creates a Copilot SDK session that explores the codebase using tools (glob, view, grep) and produces .github/copilot-instructions.md. For area-specific instructions, generateAreaInstructions() generates focused content per area, and buildAreaFrontmatter() creates YAML frontmatter with applyTo glob patterns so VS Code scopes them to the right files. These are written to .github/instructions/{sanitized-name}.instructions.md via writeAreaInstruction(). The instructions command supports --areas to generate all area instructions, --areas-only to skip the root file, and --area <name> for a single area."
     },
     {
diff --git a/docs/product.md b/docs/product.md
index ef8ece7..7ace8f4 100644
--- a/docs/product.md
+++ b/docs/product.md
@@ -10,7 +10,7 @@ This gap widens at scale. An organization with hundreds of repos can't manually
 
 - **Platform engineering teams** rolling out AI coding tools across an org — need to assess readiness, set baselines, and track adoption at scale.
 - **Individual developers** who want their AI agent to understand their repo's stack, conventions, and architecture from day one.
-- **Engineering leadership** evaluating AI readiness across portfolios, with quantifiable maturity levels and policy-driven compliance.
+- **Engineering leadership** evaluating readiness across portfolios, with quantifiable maturity levels and policy-driven compliance.
 
 ## What AgentRC Does
 
@@ -33,7 +33,7 @@ AgentRC's readiness assessment maps repos to a 5-level maturity model:
 | Level | Name         | What it means                                       |
 | ----- | ------------ | --------------------------------------------------- |
 | 1     | Functional   | Builds, tests, basic tooling in place               |
-| 2     | Documented   | README, CONTRIBUTING, custom AI instructions exist  |
+| 2     | Documented   | README, CONTRIBUTING, custom instructions exist     |
 | 3     | Standardized | CI/CD, security policies, CODEOWNERS, observability |
 | 4     | Optimized    | MCP servers, custom agents, AI skills configured    |
 | 5     | Autonomous   | Full AI-native development with minimal oversight   |
diff --git a/examples/README.md b/examples/README.md
index 51b884c..c429180 100644
--- a/examples/README.md
+++ b/examples/README.md
@@ -17,7 +17,7 @@ agentrc init /path/to/repo
 # Inspect a repository before generating config or instructions
 agentrc analyze /path/to/repo
 
-# Check AI readiness in the terminal
+# Check readiness in the terminal
 agentrc readiness /path/to/repo
 
 # Generate a visual readiness report
diff --git a/packages/core/src/services/batch.ts b/packages/core/src/services/batch.ts
index ca1b6ff..80f92b5 100644
--- a/packages/core/src/services/batch.ts
+++ b/packages/core/src/services/batch.ts
@@ -129,7 +129,7 @@ async function processRepo(params: ProcessRepoParams): Promise<ProcessResult> {
     }
 
     progress?.update(`${label}: Committing...`);
-    await commitAll(repoPath, "chore: add copilot instructions via AgentRC");
+    await commitAll(repoPath, "chore: add instructions via AgentRC");
 
     progress?.update(`${label}: Pushing...`);
     await pushBranch(repoPath, branch, token, provider);
@@ -173,7 +173,7 @@ export async function processGitHubRepo(options: ProcessGitHubRepoOptions): Prom
         token,
         owner: repo.owner,
         repo: repo.name,
-        title: "🤖 Add Copilot instructions via AgentRC",
+        title: "🤖 Add instructions via AgentRC",
         body: buildInstructionsPrBody(),
         head: branchName,
         base: repo.defaultBranch
@@ -210,7 +210,7 @@ export async function processAzureRepo(options: ProcessAzureRepoOptions): Promis
         project: repo.project,
         repoId: repo.id,
         repoName: repo.name,
-        title: "🤖 Add Copilot instructions via AgentRC",
+        title: "🤖 Add instructions via AgentRC",
         body: buildInstructionsPrBody(),
         sourceBranch: branchName,
         targetBranch: repo.defaultBranch
@@ -299,7 +299,7 @@ export type ProcessBatchReadinessRepoOptions = {
 };
 
 /**
- * Clone a single GitHub repo and run the AI-readiness report.
+ * Clone a single GitHub repo and run the readiness report.
  * Credential-stripped clone URL is restored on the remote after cloning.
  * Extracted from BatchReadinessTui so the TUI remains a thin presenter.
  */
diff --git a/packages/core/src/services/instructions.ts b/packages/core/src/services/instructions.ts
index e0dde3f..d2ded05 100644
--- a/packages/core/src/services/instructions.ts
+++ b/packages/core/src/services/instructions.ts
@@ -430,8 +430,8 @@ export async function generateAreaInstructions(
     const preferredModel = options.model ?? DEFAULT_MODEL;
 
     const areaSystemContent = hasExistingInstructions
-      ? `You are an expert codebase analyst. Your task is to generate a concise .instructions.md file for a specific area of a codebase. This file will be used as a file-based custom instruction in VS Code Copilot, automatically applied when working on files matching certain patterns. This file should complement, not duplicate, existing instruction files. Use the Explore subagents and read-only tools to explore the codebase. When done, call the emit_file_content tool with the final markdown.`
-      : `You are an expert codebase analyst. Your task is to generate a concise .instructions.md file for a specific area of a codebase. This file will be used as a file-based custom instruction in VS Code Copilot, automatically applied when working on files matching certain patterns. Use the Explore subagents and read-only tools to explore the codebase. When done, call the emit_file_content tool with the final markdown.`;
+      ? `You are an expert codebase analyst. Your task is to generate a concise .instructions.md file for a specific area of a codebase. This file will be used as an area instruction in VS Code, automatically applied when working on files matching certain patterns. This file should complement, not duplicate, existing instruction files. Use the Explore subagents and read-only tools to explore the codebase. When done, call the emit_file_content tool with the final markdown.`
+      : `You are an expert codebase analyst. Your task is to generate a concise .instructions.md file for a specific area of a codebase. This file will be used as an area instruction in VS Code, automatically applied when working on files matching certain patterns. Use the Explore subagents and read-only tools to explore the codebase. When done, call the emit_file_content tool with the final markdown.`;
 
     const { tool: emitTool, getContent } = await createEmitTool();
 
@@ -470,7 +470,7 @@ export async function generateAreaInstructions(
       }
     });
 
-    const prompt = `Analyze the "${area.name}" area of this codebase and generate a file-based instruction file.
+    const prompt = `Analyze the "${area.name}" area of this codebase and generate an area instruction file.
 
 This area covers files matching: ${applyToStr}
 ${area.description ? `Description: ${area.description}` : ""}
diff --git a/packages/core/src/services/readiness.ts b/packages/core/src/services/readiness.ts
index bdd8f9e..2932028 100644
--- a/packages/core/src/services/readiness.ts
+++ b/packages/core/src/services/readiness.ts
@@ -66,7 +66,7 @@ export function getLevelName(level: number): string {
 export function getLevelDescription(level: number): string {
   const descriptions: Record<number, string> = {
     1: "Repo builds, tests run, and basic tooling (linter, lockfile) is in place. AI agents can clone and get started.",
-    2: "README, CONTRIBUTING guide, and custom AI instructions exist. Agents understand project context and conventions.",
+    2: "README, CONTRIBUTING guide, and custom instructions exist. Agents understand project context and conventions.",
     3: "CI/CD, security policies, CODEOWNERS, and observability are configured. Agents operate within well-defined guardrails.",
     4: "MCP servers, custom agents, and AI skills are set up. Agents have deep integration with project-specific tools and workflows.",
     5: "Full AI-native development: agents can independently plan, implement, test, and ship changes with minimal human oversight."
@@ -683,7 +683,7 @@ export function buildCriteria(): ReadinessCriterion[] {
     },
     {
       id: "custom-instructions",
-      title: "Custom AI instructions or agent guidance",
+      title: "Custom instructions or agent guidance",
       pillar: "ai-tooling",
       level: 1,
       scope: "repo",
@@ -695,7 +695,7 @@ export function buildCriteria(): ReadinessCriterion[] {
           return {
             status: "fail",
             reason:
-              "Missing custom AI instructions (e.g. copilot-instructions.md, CLAUDE.md, AGENTS.md, .cursorrules).",
+              "Missing custom instructions (e.g. copilot-instructions.md, CLAUDE.md, AGENTS.md, .cursorrules).",
             evidence: [
               "copilot-instructions.md",
               "CLAUDE.md",
@@ -706,7 +706,7 @@ export function buildCriteria(): ReadinessCriterion[] {
           };
         }
 
-        // Check for file-based instructions (.github/instructions/*.instructions.md)
+        // Check for area instructions (.github/instructions/*.instructions.md)
         const fileBasedInstructions = await hasFileBasedInstructions(context.repoPath);
         const areas = context.analysis.areas ?? [];
 
@@ -715,13 +715,13 @@ export function buildCriteria(): ReadinessCriterion[] {
           if (fileBasedInstructions.length === 0) {
             return {
               status: "pass",
-              reason: `Root instructions found, but no file-based instructions for ${areas.length} detected areas. Run \`agentrc instructions --areas\` to generate.`,
+              reason: `Root instructions found, but no area instructions for ${areas.length} detected areas. Run \`agentrc instructions --areas\` to generate.`,
               evidence: [...rootFound, ...areas.map((a) => `${a.name}: missing .instructions.md`)]
             };
           }
           return {
             status: "pass",
-            reason: `Root + ${fileBasedInstructions.length} file-based instruction(s) found.`,
+            reason: `Root + ${fileBasedInstructions.length} area instruction(s) found.`,
             evidence: [...rootFound, ...fileBasedInstructions]
           };
         }
diff --git a/packages/core/src/services/readiness/criteria.ts b/packages/core/src/services/readiness/criteria.ts
index af6dfac..4043d21 100644
--- a/packages/core/src/services/readiness/criteria.ts
+++ b/packages/core/src/services/readiness/criteria.ts
@@ -278,7 +278,7 @@ export function buildCriteria(): ReadinessCriterion[] {
     },
     {
       id: "custom-instructions",
-      title: "Custom AI instructions or agent guidance",
+      title: "Custom instructions or agent guidance",
       pillar: "ai-tooling",
       level: 1,
       scope: "repo",
@@ -290,7 +290,7 @@ export function buildCriteria(): ReadinessCriterion[] {
           return {
             status: "fail",
             reason:
-              "Missing custom AI instructions (e.g. copilot-instructions.md, CLAUDE.md, AGENTS.md, .cursorrules).",
+              "Missing custom instructions (e.g. copilot-instructions.md, CLAUDE.md, AGENTS.md, .cursorrules).",
             evidence: [
               "copilot-instructions.md",
               "CLAUDE.md",
@@ -301,7 +301,7 @@ export function buildCriteria(): ReadinessCriterion[] {
           };
         }
 
-        // Check for file-based instructions (.github/instructions/*.instructions.md)
+        // Check for area instructions (.github/instructions/*.instructions.md)
         const fileBasedInstructions = await hasFileBasedInstructions(context.repoPath);
         const areas = context.analysis.areas ?? [];
 
@@ -310,13 +310,13 @@ export function buildCriteria(): ReadinessCriterion[] {
           if (fileBasedInstructions.length === 0) {
             return {
               status: "pass",
-              reason: `Root instructions found, but no file-based instructions for ${areas.length} detected areas. Run \`agentrc instructions --areas\` to generate.`,
+              reason: `Root instructions found, but no area instructions for ${areas.length} detected areas. Run \`agentrc instructions --areas\` to generate.`,
               evidence: [...rootFound, ...areas.map((a) => `${a.name}: missing .instructions.md`)]
             };
           }
           return {
             status: "pass",
-            reason: `Root + ${fileBasedInstructions.length} file-based instruction(s) found.`,
+            reason: `Root + ${fileBasedInstructions.length} area instruction(s) found.`,
             evidence: [...rootFound, ...fileBasedInstructions]
           };
         }
diff --git a/packages/core/src/services/readiness/scoring.ts b/packages/core/src/services/readiness/scoring.ts
index 3dd3bd6..bf79454 100644
--- a/packages/core/src/services/readiness/scoring.ts
+++ b/packages/core/src/services/readiness/scoring.ts
@@ -32,7 +32,7 @@ export function getLevelName(level: number): string {
 export function getLevelDescription(level: number): string {
   const descriptions: Record<number, string> = {
     1: "Repo builds, tests run, and basic tooling (linter, lockfile) is in place. AI agents can clone and get started.",
-    2: "README, CONTRIBUTING guide, and custom AI instructions exist. Agents understand project context and conventions.",
+    2: "README, CONTRIBUTING guide, and custom instructions exist. Agents understand project context and conventions.",
     3: "CI/CD, security policies, CODEOWNERS, and observability are configured. Agents operate within well-defined guardrails.",
     4: "MCP servers, custom agents, and AI skills are set up. Agents have deep integration with project-specific tools and workflows.",
     5: "Full AI-native development: agents can independently plan, implement, test, and ship changes with minimal human oversight."
diff --git a/packages/core/src/services/visualReport.ts b/packages/core/src/services/visualReport.ts
index 05340dc..e7500cd 100644
--- a/packages/core/src/services/visualReport.ts
+++ b/packages/core/src/services/visualReport.ts
@@ -8,11 +8,7 @@ type VisualReportOptions = {
 };
 
 export function generateVisualReport(options: VisualReportOptions): string {
-  const {
-    reports,
-    title = "AI Readiness Report",
-    generatedAt = new Date().toISOString()
-  } = options;
+  const { reports, title = "Readiness Report", generatedAt = new Date().toISOString() } = options;
 
   const successfulReports = reports.filter((r) => !r.error);
   const failedReports = reports.filter((r) => r.error);
@@ -610,7 +606,7 @@ export function generateVisualReport(options: VisualReportOptions): string {
     }
 
     <div class="footer">
-      <p>Generated with <a href="https://github.com/microsoft/agentrc">AgentRC</a> &middot; AI Readiness Tool</p>
+      <p>Generated with <a href="https://github.com/microsoft/agentrc">AgentRC</a> &middot; Readiness Tool</p>
     </div>
   </div>
   <script>
diff --git a/packages/core/src/utils/pr.ts b/packages/core/src/utils/pr.ts
index a3e4985..67150aa 100644
--- a/packages/core/src/utils/pr.ts
+++ b/packages/core/src/utils/pr.ts
@@ -17,9 +17,9 @@ export function isAgentrcFile(filePath: string): boolean {
 
 export function buildInstructionsPrBody(): string {
   return [
-    "## 🤖 Copilot Instructions Added",
+    "## 🤖 Instructions Added",
     "",
-    "This PR adds a `.github/copilot-instructions.md` file to help GitHub Copilot understand this codebase better.",
+    "This PR adds a `.github/copilot-instructions.md` file to help AI coding assistants understand this codebase better.",
     "",
     "### What's Included",
     "",
@@ -31,7 +31,7 @@ export function buildInstructionsPrBody(): string {
     "",
     "### Benefits",
     "",
-    "With these instructions, Copilot will:",
+    "With these instructions, AI coding assistants will:",
     "- Generate more contextually-aware code suggestions",
     "- Follow project-specific patterns and conventions",
     "- Understand the codebase structure",
@@ -51,7 +51,7 @@ export function buildFullPrBody(): string {
     "",
     "| File | Purpose |",
     "|------|---------|",
-    "| `.github/copilot-instructions.md` | Custom instructions for GitHub Copilot |",
+    "| `.github/copilot-instructions.md` | Custom instructions for AI coding assistants |",
     "| `.vscode/settings.json` | VS Code settings for optimal AI assistance |",
     "| `.vscode/mcp.json` | Model Context Protocol server configuration |",
     "",
@@ -65,7 +65,7 @@ export function buildFullPrBody(): string {
     "",
     "### Benefits",
     "",
-    "With these configurations, Copilot will:",
+    "With these configurations, AI coding assistants will:",
     "- Generate more contextually-aware code suggestions",
     "- Follow project-specific patterns and conventions",
     "- Understand the codebase structure",
@@ -75,7 +75,7 @@ export function buildFullPrBody(): string {
     "",
     "1. Merge this PR",
     "2. Open the project in VS Code",
-    "3. Start chatting with Copilot — it now understands your project!",
+    "3. Start using AI assistants — they now understand your project!",
     "",
     "---",
     "*Generated by [AgentRC](https://github.com/microsoft/agentrc) - Prime your repos for AI*"
diff --git a/src/cli.ts b/src/cli.ts
index b3867ac..f0b11c1 100644
--- a/src/cli.ts
+++ b/src/cli.ts
@@ -53,7 +53,7 @@ export function runCli(argv: string[]): void {
 
   program
     .command("init")
-    .description("Interactive repo setup — analyze, generate instructions and configs")
+    .description("Init repository — analyze & generate instructions")
     .argument("[path]", "Path to a local repository")
     .option("--github", "Use a GitHub repository")
     .option("--provider <provider>", "Repo provider (github|azure)")
@@ -83,7 +83,7 @@ export function runCli(argv: string[]): void {
     )
     .argument("[path]", "Path to a local repository")
     .option("--force", "Overwrite existing files")
-    .option("--per-app", "Generate per-app in monorepos")
+    .option("--per-app", "(deprecated) Use `agentrc instructions --areas` instead")
     .option("--model <name>", "Model for instructions generation", DEFAULT_MODEL)
     .option("--strategy <mode>", "Instruction strategy (flat or nested)")
     .action(withGlobalOpts(generateCommand));
@@ -120,21 +120,22 @@ export function runCli(argv: string[]): void {
 
   program
     .command("instructions")
-    .description("Generate root and per-area AI instruction files")
+    .description("Generate instructions for the repository")
     .option("--repo <path>", "Repository path", process.cwd())
-    .option("--output <path>", "Output path for copilot instructions")
+    .option("--output <path>", "Output path for instructions")
     .option("--model <name>", "Model for instructions generation", DEFAULT_MODEL)
     .option("--force", "Overwrite existing area instruction files")
-    .option("--areas", "Also generate file-based instructions for detected areas")
-    .option("--areas-only", "Generate only file-based area instructions (skip root)")
-    .option("--area <name>", "Generate file-based instructions for a specific area")
+    .option("--areas", "Also generate instructions for detected areas")
+    .option("--areas-only", "Generate only area instructions (skip root)")
+    .option("--area <name>", "Generate instructions for a specific area")
     .option("--strategy <mode>", "Instruction strategy (flat or nested)")
     .option("--claude-md", "Generate CLAUDE.md files alongside AGENTS.md (nested strategy)")
+    .option("--dry-run", "Preview generated files without writing anything")
     .action(withGlobalOpts(instructionsCommand));
 
   program
     .command("readiness")
-    .description("AI readiness assessment across 9 maturity pillars")
+    .description("Run readiness report across 9 maturity pillars")
     .argument("[path]", "Path to a local repository")
     .option("--output <path>", "Write report to file (.json, .md, or .html)")
     .option("--force", "Overwrite existing output file")
@@ -156,7 +157,7 @@ export function runCli(argv: string[]): void {
 
   program
     .command("batch-readiness")
-    .description("Generate batch AI readiness report for multiple repos")
+    .description("Run batch readiness report for multiple repos")
     .option("--output <path>", "Write HTML report to file")
     .option("--policy <sources>", "Policy sources (comma-separated: paths, npm packages)")
     .action(withGlobalOpts(batchReadinessCommand));
diff --git a/src/commands/generate.ts b/src/commands/generate.ts
index 365ccbb..43fa837 100644
--- a/src/commands/generate.ts
+++ b/src/commands/generate.ts
@@ -25,6 +25,16 @@ export async function generateCommand(
   const repoPath = path.resolve(repoPathArg ?? process.cwd());
 
   if (type === "instructions" || type === "agents") {
+    if (!options.quiet) {
+      process.stderr.write(
+        `⚠ \`generate ${type}\` is deprecated — use \`agentrc instructions\` directly.\n`
+      );
+    }
+    if (options.perApp && !options.quiet) {
+      process.stderr.write(
+        `⚠ --per-app is deprecated — use \`agentrc instructions --areas\` instead.\n`
+      );
+    }
     // Delegate to the canonical instructions command
     const output = type === "agents" ? path.join(repoPath, "AGENTS.md") : undefined;
     await instructionsCommand({
diff --git a/src/commands/init.ts b/src/commands/init.ts
index 7f6e40e..7f44ca1 100644
--- a/src/commands/init.ts
+++ b/src/commands/init.ts
@@ -283,7 +283,7 @@ export async function initCommand(
       process.stderr.write(`${file.action === "wrote" ? "Wrote" : "Skipped"} ${file.path}\n`);
     }
     process.stderr.write("\nNext steps:\n");
-    process.stderr.write("  agentrc readiness             Check AI readiness across 9 pillars\n");
+    process.stderr.write("  agentrc readiness             Run readiness report across 9 pillars\n");
     if (analysis.areas && analysis.areas.length > 0) {
       process.stderr.write("  agentrc instructions --areas   Generate per-area instructions\n");
     }
diff --git a/src/commands/instructions.ts b/src/commands/instructions.ts
index 895d73c..e7a1297 100644
--- a/src/commands/instructions.ts
+++ b/src/commands/instructions.ts
@@ -7,6 +7,7 @@ import {
   generateAreaInstructions,
   generateNestedInstructions,
   generateNestedAreaInstructions,
+  areaInstructionPath,
   writeAreaInstruction,
   writeNestedInstructions
 } from "@agentrc/core/services/instructions";
@@ -37,6 +38,7 @@ type InstructionsOptions = {
   area?: string;
   strategy?: string;
   claudeMd?: boolean;
+  dryRun?: boolean;
 };
 
 export async function instructionsCommand(options: InstructionsOptions): Promise<void> {
@@ -71,6 +73,8 @@ export async function instructionsCommand(options: InstructionsOptions): Promise
   }
 
   try {
+    const dryRunFiles: { path: string; bytes: number }[] = [];
+
     // Generate root instructions unless --areas-only
     if (!options.areasOnly && !options.area) {
       if (strategy === "nested") {
@@ -84,25 +88,59 @@ export async function instructionsCommand(options: InstructionsOptions): Promise
             detailDir,
             claudeMd
           });
-          const actions = await writeNestedInstructions(repoPath, nestedResult, options.force);
-          for (const action of actions) {
-            const relPath = path.relative(process.cwd(), action.path);
-            if (action.action === "wrote") {
-              if (shouldLog(options)) progress.succeed(`Wrote ${relPath}`);
-            } else if (shouldLog(options)) {
-              progress.update(`Skipped ${relPath} (${skipReason(action.action)})`);
+          if (options.dryRun) {
+            const dryFiles = [
+              { path: nestedResult.hub.relativePath, content: nestedResult.hub.content },
+              ...nestedResult.details.map((d) => ({ path: d.relativePath, content: d.content })),
+              ...(nestedResult.claudeMd
+                ? [
+                    {
+                      path: nestedResult.claudeMd.relativePath,
+                      content: nestedResult.claudeMd.content
+                    }
+                  ]
+                : [])
+            ];
+            for (const file of dryFiles) {
+              const relPath = path.relative(process.cwd(), path.join(repoPath, file.path));
+              if (shouldLog(options)) {
+                progress.update(
+                  `[dry-run] Would write ${relPath} (${Buffer.byteLength(file.content, "utf8")} bytes)`
+                );
+              }
+            }
+            if (options.json) {
+              dryRunFiles.push(
+                ...dryFiles.map((f) => ({
+                  path: f.path,
+                  bytes: Buffer.byteLength(f.content, "utf8")
+                }))
+              );
+            }
+            for (const warning of nestedResult.warnings) {
+              if (shouldLog(options)) progress.update(`Warning: ${warning}`);
+            }
+          } else {
+            const actions = await writeNestedInstructions(repoPath, nestedResult, options.force);
+            for (const action of actions) {
+              const relPath = path.relative(process.cwd(), action.path);
+              if (action.action === "wrote") {
+                if (shouldLog(options)) progress.succeed(`Wrote ${relPath}`);
+              } else if (shouldLog(options)) {
+                progress.update(`Skipped ${relPath} (${skipReason(action.action)})`);
+              }
+            }
+            for (const warning of nestedResult.warnings) {
+              if (shouldLog(options)) progress.update(`Warning: ${warning}`);
+            }
+            if (options.json) {
+              const result: CommandResult<{ files: typeof actions }> = {
+                ok: true,
+                status: "success",
+                data: { files: actions }
+              };
+              outputResult(result, true);
             }
-          }
-          for (const warning of nestedResult.warnings) {
-            if (shouldLog(options)) progress.update(`Warning: ${warning}`);
-          }
-          if (options.json) {
-            const result: CommandResult<{ files: typeof actions }> = {
-              ok: true,
-              status: "success",
-              data: { files: actions }
-            };
-            outputResult(result, true);
           }
         } catch (error) {
           const msg =
@@ -134,42 +172,55 @@ export async function instructionsCommand(options: InstructionsOptions): Promise
         }
 
         if (content) {
-          await ensureDir(path.dirname(outputPath));
-          const { wrote, reason } = await safeWriteFile(
-            outputPath,
-            content,
-            Boolean(options.force)
-          );
-
-          if (!wrote) {
-            const relPath = path.relative(process.cwd(), outputPath);
-            const why = reason === "symlink" ? "path is a symlink" : "file exists (use --force)";
+          if (options.dryRun) {
+            const relPath = path.relative(repoPath, outputPath);
+            const displayPath = path.relative(process.cwd(), outputPath);
+            const byteCount = Buffer.byteLength(content, "utf8");
+            if (shouldLog(options)) {
+              progress.update(`[dry-run] Would write ${displayPath} (${byteCount} bytes)`);
+            }
             if (options.json) {
-              const result: CommandResult<{ outputPath: string; skipped: true; reason: string }> = {
-                ok: true,
-                status: "noop",
-                data: { outputPath, skipped: true, reason: why }
-              };
-              outputResult(result, true);
-            } else if (shouldLog(options)) {
-              progress.update(`Skipped ${relPath}: ${why}`);
+              dryRunFiles.push({ path: relPath, bytes: byteCount });
             }
           } else {
-            const byteCount = Buffer.byteLength(content, "utf8");
+            await ensureDir(path.dirname(outputPath));
+            const { wrote, reason } = await safeWriteFile(
+              outputPath,
+              content,
+              Boolean(options.force)
+            );
 
-            if (options.json) {
-              const result: CommandResult<{
-                outputPath: string;
-                model: string;
-                byteCount: number;
-              }> = {
-                ok: true,
-                status: "success",
-                data: { outputPath, model: options.model ?? "default", byteCount }
-              };
-              outputResult(result, true);
-            } else if (shouldLog(options)) {
-              progress.succeed(`Updated ${path.relative(process.cwd(), outputPath)}`);
+            if (!wrote) {
+              const relPath = path.relative(process.cwd(), outputPath);
+              const why = reason === "symlink" ? "path is a symlink" : "file exists (use --force)";
+              if (options.json) {
+                const result: CommandResult<{ outputPath: string; skipped: true; reason: string }> =
+                  {
+                    ok: true,
+                    status: "noop",
+                    data: { outputPath, skipped: true, reason: why }
+                  };
+                outputResult(result, true);
+              } else if (shouldLog(options)) {
+                progress.update(`Skipped ${relPath}: ${why}`);
+              }
+            } else {
+              const byteCount = Buffer.byteLength(content, "utf8");
+
+              if (options.json) {
+                const result: CommandResult<{
+                  outputPath: string;
+                  model: string;
+                  byteCount: number;
+                }> = {
+                  ok: true,
+                  status: "success",
+                  data: { outputPath, model: options.model ?? "default", byteCount }
+                };
+                outputResult(result, true);
+              } else if (shouldLog(options)) {
+                progress.succeed(`Updated ${path.relative(process.cwd(), outputPath)}`);
+              }
             }
           }
         }
@@ -211,7 +262,7 @@ export async function instructionsCommand(options: InstructionsOptions): Promise
       }
 
       if (shouldLog(options)) {
-        progress.update(`Generating file-based instructions for ${targetAreas.length} area(s)...`);
+        progress.update(`Generating instructions for ${targetAreas.length} area(s)...`);
       }
 
       for (const area of targetAreas) {
@@ -234,17 +285,48 @@ export async function instructionsCommand(options: InstructionsOptions): Promise
               detailDir,
               claudeMd
             });
-            const actions = await writeNestedInstructions(repoPath, nestedResult, options.force);
-            for (const action of actions) {
-              const relPath = path.relative(process.cwd(), action.path);
-              if (action.action === "wrote") {
-                if (shouldLog(options)) progress.succeed(`Wrote ${relPath}`);
-              } else if (shouldLog(options)) {
-                progress.update(`Skipped ${relPath} (${skipReason(action.action)})`);
+            if (options.dryRun) {
+              const dryFiles = [
+                { path: nestedResult.hub.relativePath, content: nestedResult.hub.content },
+                ...nestedResult.details.map((d) => ({ path: d.relativePath, content: d.content })),
+                ...(nestedResult.claudeMd
+                  ? [
+                      {
+                        path: nestedResult.claudeMd.relativePath,
+                        content: nestedResult.claudeMd.content
+                      }
+                    ]
+                  : [])
+              ];
+              for (const file of dryFiles) {
+                const relPath = path.relative(process.cwd(), path.join(repoPath, file.path));
+                if (shouldLog(options)) {
+                  progress.update(
+                    `[dry-run] Would write ${relPath} (${Buffer.byteLength(file.content, "utf8")} bytes)`
+                  );
+                }
+              }
+              if (options.json) {
+                dryRunFiles.push(
+                  ...dryFiles.map((f) => ({
+                    path: f.path,
+                    bytes: Buffer.byteLength(f.content, "utf8")
+                  }))
+                );
+              }
+            } else {
+              const actions = await writeNestedInstructions(repoPath, nestedResult, options.force);
+              for (const action of actions) {
+                const relPath = path.relative(process.cwd(), action.path);
+                if (action.action === "wrote") {
+                  if (shouldLog(options)) progress.succeed(`Wrote ${relPath}`);
+                } else if (shouldLog(options)) {
+                  progress.update(`Skipped ${relPath} (${skipReason(action.action)})`);
+                }
+              }
+              for (const warning of nestedResult.warnings) {
+                if (shouldLog(options)) progress.update(`Warning: ${warning}`);
               }
-            }
-            for (const warning of nestedResult.warnings) {
-              if (shouldLog(options)) progress.update(`Warning: ${warning}`);
             }
           } else {
             // Flat: existing behavior
@@ -262,21 +344,37 @@ export async function instructionsCommand(options: InstructionsOptions): Promise
               continue;
             }
 
-            const result = await writeAreaInstruction(repoPath, area, body, options.force);
-            if (result.status === "skipped") {
+            if (options.dryRun) {
               if (shouldLog(options)) {
-                progress.update(`Skipped "${area.name}" — file exists (use --force to overwrite).`);
+                progress.update(
+                  `[dry-run] Would write area "${area.name}" (${Buffer.byteLength(body, "utf8")} bytes)`
+                );
+              }
+              if (options.json) {
+                dryRunFiles.push({
+                  path: path.relative(repoPath, areaInstructionPath(repoPath, area)),
+                  bytes: Buffer.byteLength(body, "utf8")
+                });
+              }
+            } else {
+              const result = await writeAreaInstruction(repoPath, area, body, options.force);
+              if (result.status === "skipped") {
+                if (shouldLog(options)) {
+                  progress.update(
+                    `Skipped "${area.name}" — file exists (use --force to overwrite).`
+                  );
+                }
+                continue;
+              }
+              if (result.status === "symlink") {
+                if (shouldLog(options)) {
+                  progress.update(`Skipped "${area.name}" — path is a symlink.`);
+                }
+                continue;
               }
-              continue;
-            }
-            if (result.status === "symlink") {
               if (shouldLog(options)) {
-                progress.update(`Skipped "${area.name}" — path is a symlink.`);
+                progress.succeed(`Wrote ${path.relative(process.cwd(), result.filePath)}`);
               }
-              continue;
-            }
-            if (shouldLog(options)) {
-              progress.succeed(`Wrote ${path.relative(process.cwd(), result.filePath)}`);
             }
           }
         } catch (error) {
@@ -289,6 +387,13 @@ export async function instructionsCommand(options: InstructionsOptions): Promise
       }
     }
 
+    if (options.dryRun && options.json) {
+      outputResult(
+        { ok: true, status: "noop" as const, data: { dryRun: true, files: dryRunFiles } },
+        true
+      );
+    }
+
     if (!wantAreas && shouldLog(options) && !options.json) {
       process.stderr.write("\nNext steps:\n");
       process.stderr.write("  agentrc eval --init      Scaffold evaluation test cases\n");
diff --git a/src/commands/readiness.ts b/src/commands/readiness.ts
index f0e6a8d..1f4c8df 100644
--- a/src/commands/readiness.ts
+++ b/src/commands/readiness.ts
@@ -100,7 +100,7 @@ export async function readinessCommand(
   if (options.visual || outputExt === ".html") {
     const html = generateVisualReport({
       reports: [{ repo: repoName, report }],
-      title: `AI Readiness Report: ${repoName}`,
+      title: `Readiness Report: ${repoName}`,
       generatedAt: new Date().toISOString()
     });
 
@@ -215,7 +215,7 @@ function printReadinessChecklist(report: ReadinessReport): void {
   }
 
   if (report.extras.length) {
-    log(chalk.bold("\nAI readiness extras"));
+    log(chalk.bold("\nReadiness extras"));
     for (const extra of report.extras) {
       const icon = extra.status === "pass" ? chalk.green("✔") : chalk.red("✖");
       log(`${icon} ${extra.title}`);
@@ -295,7 +295,7 @@ function printAreaBreakdown(areaReports: AreaReadinessReport[]): void {
 export function formatReadinessMarkdown(report: ReadinessReport, repoName: string): string {
   const lines: string[] = [];
 
-  lines.push(`# AI Readiness Report: ${repoName}`);
+  lines.push(`# Readiness Report: ${repoName}`);
   lines.push("");
   lines.push(`**Level ${report.achievedLevel}** — ${levelName(report.achievedLevel)}`);
   lines.push("");
@@ -336,7 +336,7 @@ export function formatReadinessMarkdown(report: ReadinessReport, repoName: strin
 
   // Extras
   if (report.extras.length > 0) {
-    lines.push("## AI Readiness Extras");
+    lines.push("## Readiness Extras");
     lines.push("");
     for (const extra of report.extras) {
       const icon = extra.status === "pass" ? "✅" : "❌";
diff --git a/src/services/__tests__/readiness-markdown.test.ts b/src/services/__tests__/readiness-markdown.test.ts
index 45584d7..b80a9dc 100644
--- a/src/services/__tests__/readiness-markdown.test.ts
+++ b/src/services/__tests__/readiness-markdown.test.ts
@@ -59,14 +59,14 @@ describe("formatReadinessMarkdown", () => {
         },
         {
           id: "custom-instructions",
-          title: "Custom AI instructions",
+          title: "Custom instructions",
           pillar: "ai-tooling",
           level: 1,
           scope: "repo",
           impact: "high",
           effort: "low",
           status: "fail",
-          reason: "Missing custom AI instructions."
+          reason: "Missing custom instructions."
         }
       ],
       extras: [
@@ -79,7 +79,7 @@ describe("formatReadinessMarkdown", () => {
 
   it("renders heading with repo name and level", () => {
     const md = formatReadinessMarkdown(makeReport(), "my-repo");
-    expect(md).toContain("# AI Readiness Report: my-repo");
+    expect(md).toContain("# Readiness Report: my-repo");
     expect(md).toContain("**Level 1** — Functional");
   });
 
@@ -112,12 +112,12 @@ describe("formatReadinessMarkdown", () => {
     const md = formatReadinessMarkdown(makeReport(), "my-repo");
     expect(md).toContain("## Fix First");
     expect(md).toContain("README present");
-    expect(md).toContain("Custom AI instructions");
+    expect(md).toContain("Custom instructions");
   });
 
   it("includes extras section", () => {
     const md = formatReadinessMarkdown(makeReport(), "my-repo");
-    expect(md).toContain("## AI Readiness Extras");
+    expect(md).toContain("## Readiness Extras");
     expect(md).toContain("✅ AGENTS.md present");
     expect(md).toContain("❌ Architecture guide present");
   });
diff --git a/src/services/__tests__/readiness-output.test.ts b/src/services/__tests__/readiness-output.test.ts
index 1b8e532..244c748 100644
--- a/src/services/__tests__/readiness-output.test.ts
+++ b/src/services/__tests__/readiness-output.test.ts
@@ -44,7 +44,7 @@ describe("readinessCommand --output", () => {
     await readinessCommand(repoPath, { output: outputPath, quiet: true });
 
     const content = await fs.readFile(outputPath, "utf-8");
-    expect(content).toContain("# AI Readiness Report:");
+    expect(content).toContain("# Readiness Report:");
     expect(content).toContain("## Repo Health");
   });
 
@@ -186,7 +186,7 @@ describe("readinessCommand --output", () => {
     expect(parsed.data).toBeDefined();
 
     const fileContent = await fs.readFile(outputPath, "utf-8");
-    expect(fileContent).toContain("# AI Readiness Report:");
+    expect(fileContent).toContain("# Readiness Report:");
   });
 
   it("emits JSON to stdout when --json is used with html output", async () => {
diff --git a/src/services/__tests__/readiness.test.ts b/src/services/__tests__/readiness.test.ts
index 2925c49..43a84c3 100644
--- a/src/services/__tests__/readiness.test.ts
+++ b/src/services/__tests__/readiness.test.ts
@@ -230,7 +230,7 @@ describe("runReadinessReport", () => {
       expect(criterion?.status).toBe("pass");
     });
 
-    it("mentions missing file-based instructions when areas detected", async () => {
+    it("mentions missing area instructions when areas detected", async () => {
       await writePackageJson({ name: "test-repo" });
       await writeFile(".github/copilot-instructions.md", "# Instructions");
       // Create a heuristic area directory
@@ -240,18 +240,18 @@ describe("runReadinessReport", () => {
       const criterion = report.criteria.find((c) => c.id === "custom-instructions");
 
       expect(criterion?.status).toBe("pass");
-      expect(criterion?.reason).toContain("no file-based instructions");
+      expect(criterion?.reason).toContain("no area instructions");
       expect(criterion?.evidence).toEqual(
         expect.arrayContaining([expect.stringContaining("missing .instructions.md")])
       );
     });
 
-    it("reports file-based instructions count when present with areas", async () => {
+    it("reports area instructions count when present with areas", async () => {
       await writePackageJson({ name: "test-repo" });
       await writeFile(".github/copilot-instructions.md", "# Instructions");
       // Create a heuristic area directory
       await writeFile("frontend/index.ts", "export {};");
-      // Create a file-based instruction
+      // Create an area instruction
       await writeFile(
         ".github/instructions/frontend.instructions.md",
         "---\napplyTo: frontend/**\n---\n# Frontend"
@@ -261,7 +261,7 @@ describe("runReadinessReport", () => {
       const criterion = report.criteria.find((c) => c.id === "custom-instructions");
 
       expect(criterion?.status).toBe("pass");
-      expect(criterion?.reason).toContain("file-based instruction");
+      expect(criterion?.reason).toContain("area instruction");
       expect(criterion?.evidence).toEqual(
         expect.arrayContaining([expect.stringContaining("frontend.instructions.md")])
       );
diff --git a/src/services/__tests__/visualReport.test.ts b/src/services/__tests__/visualReport.test.ts
index 6252936..d2ffda7 100644
--- a/src/services/__tests__/visualReport.test.ts
+++ b/src/services/__tests__/visualReport.test.ts
@@ -56,7 +56,7 @@ function makeReport(overrides: Partial<ReadinessReport> = {}): ReadinessReport {
       },
       {
         id: "custom-instructions",
-        title: "Custom AI instructions",
+        title: "Custom instructions",
         pillar: "ai-tooling",
         level: 1,
         scope: "repo",
diff --git a/src/ui/BatchReadinessTui.tsx b/src/ui/BatchReadinessTui.tsx
index d9b8a03..84a5fd4 100644
--- a/src/ui/BatchReadinessTui.tsx
+++ b/src/ui/BatchReadinessTui.tsx
@@ -150,7 +150,7 @@ export function BatchReadinessTui({ token, outputPath, policies }: Props): React
             },
             error: r.error
           })),
-        title: "Batch AI Readiness Report",
+        title: "Batch Readiness Report",
         generatedAt: new Date().toISOString()
       });
 
diff --git a/src/ui/tui.tsx b/src/ui/tui.tsx
index 099efea..6f8a48d 100644
--- a/src/ui/tui.tsx
+++ b/src/ui/tui.tsx
@@ -2,7 +2,7 @@ import fs from "fs/promises";
 import path from "path";
 
 import type { RepoApp, Area } from "@agentrc/core/services/analyzer";
-import { analyzeRepo } from "@agentrc/core/services/analyzer";
+import { analyzeRepo, loadAgentrcConfig } from "@agentrc/core/services/analyzer";
 import { getAzureDevOpsToken } from "@agentrc/core/services/azureDevops";
 import { listCopilotModels } from "@agentrc/core/services/copilot";
 import { generateEvalScaffold } from "@agentrc/core/services/evalScaffold";
@@ -16,6 +16,7 @@ import {
   areaInstructionPath,
   writeAreaInstruction
 } from "@agentrc/core/services/instructions";
+import { runReadinessReport, getLevelName } from "@agentrc/core/services/readiness";
 import { safeWriteFile, buildTimestampedName } from "@agentrc/core/utils/fs";
 import { Box, Text, useApp, useInput, useStdout, useIsScreenReaderEnabled } from "ink";
 import type { Key } from "ink";
@@ -36,6 +37,7 @@ type Status =
   | "generating"
   | "bootstrapping"
   | "evaluating"
+  | "readiness-running"
   | "preview"
   | "done"
   | "error"
@@ -191,13 +193,15 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
   const [isMonorepo, setIsMonorepo] = useState(false);
   const [repoAreas, setRepoAreas] = useState<Area[]>([]);
   const [areaCursor, setAreaCursor] = useState(0);
+
   const repoLabel = useMemo(() => path.basename(repoPath), [repoPath]);
   const repoFull = useMemo(() => repoPath, [repoPath]);
   const isLoading =
     status === "generating" ||
     status === "bootstrapping" ||
     status === "evaluating" ||
-    status === "generating-areas";
+    status === "generating-areas" ||
+    status === "readiness-running";
   const isMenu =
     status === "model-pick" ||
     status === "eval-pick" ||
@@ -479,14 +483,14 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
               }
               return;
             }
-            if (input.toLowerCase() === "f") {
+            if (input.toLowerCase() === "n") {
               if (repoAreas.length === 0) {
                 setMessage("No areas detected. Add agentrc.config.json to define areas.");
                 return;
               }
               setAreaCursor(0);
               setStatus("generate-area-pick");
-              setMessage("Generate file-based instructions for areas.");
+              setMessage("Generate nested instructions for areas.");
               return;
             }
             if (key.escape) {
@@ -564,10 +568,7 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
             if (input.toLowerCase() === "a") {
               // All areas
               setStatus("generating-areas");
-              addLog(
-                `Generating file-based instructions for ${repoAreas.length} areas...`,
-                "progress"
-              );
+              addLog(`Generating nested instructions for ${repoAreas.length} areas...`, "progress");
               let written = 0;
               for (const [i, area] of repoAreas.entries()) {
                 setMessage(`Generating for "${area.name}" (${i + 1}/${repoAreas.length})...`);
@@ -590,9 +591,7 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
                 }
               }
               setStatus("done");
-              setMessage(
-                `Generated file-based instructions for ${written}/${repoAreas.length} areas.`
-              );
+              setMessage(`Generated nested instructions for ${written}/${repoAreas.length} areas.`);
               return;
             }
             if (key.upArrow) {
@@ -608,7 +607,7 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
               if (!area) return;
               setStatus("generating-areas");
               setMessage(`Generating for "${area.name}"...`);
-              addLog(`Generating file-based instructions for "${area.name}"...`, "progress");
+              addLog(`Generating nested instructions for "${area.name}"...`, "progress");
               try {
                 const body = await generateAreaInstructions({
                   repoPath,
@@ -789,6 +788,38 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
               return;
             }
 
+            if (input.toLowerCase() === "r") {
+              setStatus("readiness-running");
+              setMessage("Running readiness report…");
+              addLog("Running readiness report…", "progress");
+              try {
+                let policies: string[] | undefined;
+                try {
+                  const config = await loadAgentrcConfig(repoPath);
+                  policies = config?.policies;
+                } catch {
+                  // Non-fatal
+                }
+                const report = await runReadinessReport({ repoPath, policies });
+                const levelName = getLevelName(report.achievedLevel);
+                const failCount = report.criteria.filter((c) => c.status === "fail").length;
+                addLog(
+                  `Level ${report.achievedLevel} (${levelName}) — ${failCount} item(s) to fix`,
+                  "success"
+                );
+                setStatus("done");
+                setMessage(
+                  `Readiness: Level ${report.achievedLevel} (${levelName}). ${failCount} improvement(s) available.`
+                );
+              } catch (error) {
+                const msg = error instanceof Error ? error.message : "Failed.";
+                addLog(`Readiness failed: ${msg}`, "error");
+                setStatus("error");
+                setMessage(`Readiness failed: ${msg}`);
+              }
+              return;
+            }
+
             if (input.toLowerCase() === "b") {
               setStatus("batch-pick");
               setMessage("Select batch provider.");
@@ -874,7 +905,9 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
                 ? "batch"
                 : status === "model-pick"
                   ? "models"
-                  : status;
+                  : status === "readiness-running"
+                    ? "readiness"
+                    : status;
   const statusColor =
     status === "error"
       ? "red"
@@ -1090,14 +1123,10 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
         </>
       )}
 
-      {/* Area picker for file-based instructions */}
+      {/* Area picker for nested instructions */}
       {status === "generate-area-pick" && repoAreas.length > 0 && (
         <>
-          <Divider
-            columns={terminalColumns}
-            label="File-based instructions"
-            accessible={accessible}
-          />
+          <Divider columns={terminalColumns} label="Nested instructions" accessible={accessible} />
           <Box flexDirection="column" paddingLeft={1}>
             {repoAreas.map((area, i) => (
               <Text key={area.name}>
@@ -1215,9 +1244,9 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
           </Box>
         ) : status === "generate-pick" ? (
           <Box>
-            <KeyHint k="C" label="Copilot instructions" />
-            <KeyHint k="A" label="AGENTS.md" />
-            {repoAreas.length > 0 && <KeyHint k="F" label="File-based (areas)" />}
+            <KeyHint k="C" label="Instructions" />
+            <KeyHint k="A" label="Agents" />
+            {repoAreas.length > 0 && <KeyHint k="N" label="Nested (areas)" />}
             <KeyHint k="Esc" label="Back" />
           </Box>
         ) : status === "generate-app-pick" ? (
@@ -1278,6 +1307,7 @@ export function AgentRCTui({ repoPath, skipAnimation = false }: Props): React.JS
           <Box flexDirection="column">
             <Box>
               <KeyHint k="G" label="Generate" />
+              <KeyHint k="R" label="Readiness" />
               <KeyHint k="E" label="Eval" />
               <KeyHint k="B" label="Batch" />
             </Box>
diff --git a/vscode-extension/README.md b/vscode-extension/README.md
index de8adef..a5c148f 100644
--- a/vscode-extension/README.md
+++ b/vscode-extension/README.md
@@ -16,7 +16,7 @@ Detect languages, frameworks, package managers, and monorepo structure. Results
 
 `AgentRC: Analyze Repository`
 
-### AI Readiness Assessment
+### Readiness Report
 
 Score your repo across **9 pillars** grouped into **Repo Health** and **AI Setup**, with maturity levels from Functional (1) to Autonomous (5).
 
@@ -24,7 +24,7 @@ Score your repo across **9 pillars** grouped into **Repo Health** and **AI Setup
 - Drill-down into criteria in the **Readiness** tree view
 - Pass/fail icons with evidence for each criterion
 
-`AgentRC: AI Readiness Report`
+`AgentRC: Readiness Report`
 
 ### Generate Instructions
 
@@ -35,7 +35,11 @@ Create AI instruction files using the Copilot SDK. Choose your format:
 
 For monorepos, pick specific areas to generate per-area instruction files with `applyTo` scoping.
 
-`AgentRC: Generate Copilot Instructions`
+`AgentRC: Generate Instructions`
+
+In multi-root workspaces, generate instructions for all workspace roots at once.
+
+`AgentRC: Generate Instructions (All Roots)`
 
 ### Generate Configs
 
@@ -76,12 +80,13 @@ Both views show welcome screens with action buttons when no data is loaded yet.
 
 | Setting               | Default             | Description                                        |
 | --------------------- | ------------------- | -------------------------------------------------- |
-| `agentrc.model`       | `claude-sonnet-4.5` | Default Copilot model for generation               |
+| `agentrc.model`       | `claude-sonnet-4.6` | Default Copilot model for generation               |
 | `agentrc.autoAnalyze` | `false`             | Automatically analyze repository on workspace open |
+| `agentrc.judgeModel`  | _(uses model)_      | Copilot model for judging eval responses           |
 
 ## Requirements
 
-- **VS Code 1.100.0+**
+- **VS Code 1.109.0+**
 - **GitHub Copilot Chat extension** (provides the Copilot CLI)
 - **Copilot authentication** — run `copilot` → `/login` in your terminal
 - **GitHub account** — for GitHub PR creation (authenticated via VS Code)
diff --git a/vscode-extension/package.json b/vscode-extension/package.json
index 4194275..3a9e1eb 100644
--- a/vscode-extension/package.json
+++ b/vscode-extension/package.json
@@ -1,7 +1,7 @@
 {
   "name": "agentrc-vscode",
   "displayName": "AgentRC — AI Repository Setup",
-  "description": "Prime repositories for AI-assisted development: generate Copilot instructions, VS Code configs, MCP configs, and evaluate AI agent effectiveness.",
+  "description": "Prime repositories for AI-assisted development: generate instructions, VS Code configs, MCP configs, and evaluate AI agent effectiveness.",
   "version": "0.0.2",
   "publisher": "ms-vscode",
   "license": "MIT",
@@ -42,13 +42,13 @@
       },
       {
         "command": "agentrc.instructions",
-        "title": "Generate Copilot Instructions",
+        "title": "Generate Instructions",
         "category": "AgentRC",
         "icon": "$(edit)"
       },
       {
         "command": "agentrc.readiness",
-        "title": "AI Readiness Report",
+        "title": "Readiness Report",
         "category": "AgentRC",
         "icon": "$(shield)"
       },
@@ -66,7 +66,7 @@
       },
       {
         "command": "agentrc.init",
-        "title": "Initialize Repository",
+        "title": "Init Repository",
         "category": "AgentRC",
         "icon": "$(rocket)"
       },
@@ -75,6 +75,12 @@
         "title": "Create Pull Request",
         "category": "AgentRC",
         "icon": "$(git-pull-request)"
+      },
+      {
+        "command": "agentrc.batchInstructions",
+        "title": "Generate Instructions (All Roots)",
+        "category": "AgentRC",
+        "icon": "$(files)"
       }
     ],
     "viewsContainers": {
@@ -113,7 +119,7 @@
       },
       {
         "view": "agentrc.readiness",
-        "contents": "Score your repo across 9 AI-readiness pillars — from build systems to custom instructions.\n[$(shield) Run Assessment](command:agentrc.readiness)",
+        "contents": "Run readiness report — score your repo across 9 maturity pillars.\n[$(shield) Run Readiness Report](command:agentrc.readiness)",
         "when": "workspaceFolderCount > 0"
       },
       {
@@ -141,7 +147,7 @@
           },
           {
             "id": "instructions",
-            "title": "Generate AI Instructions",
+            "title": "Generate Instructions",
             "description": "Create custom instructions that teach AI about your codebase.\n[$(edit) Generate Instructions](command:agentrc.instructions)",
             "media": {
               "markdown": "resources/walkthrough/instructions.md"
@@ -163,8 +169,8 @@
           },
           {
             "id": "readiness",
-            "title": "Check AI Readiness",
-            "description": "Score your repo across 9 pillars with a visual maturity report.\n[$(shield) Run Assessment](command:agentrc.readiness)",
+            "title": "Run Readiness Report",
+            "description": "Run readiness report — score your repo across 9 maturity pillars.\n[$(shield) Run Readiness Report](command:agentrc.readiness)",
             "media": {
               "markdown": "resources/walkthrough/readiness.md"
             },
@@ -238,6 +244,10 @@
         {
           "command": "agentrc.pr",
           "when": "workspaceFolderCount > 0"
+        },
+        {
+          "command": "agentrc.batchInstructions",
+          "when": "workspaceFolderCount > 1"
         }
       ],
       "view/title": [
diff --git a/vscode-extension/resources/walkthrough/readiness.md b/vscode-extension/resources/walkthrough/readiness.md
index e719b38..76cbdd2 100644
--- a/vscode-extension/resources/walkthrough/readiness.md
+++ b/vscode-extension/resources/walkthrough/readiness.md
@@ -1,4 +1,4 @@
-## How AI-Ready Is Your Repo?
+## How Ready Is Your Repo?
 
 AgentRC evaluates your repository across **9 pillars** organized into two groups:
 
diff --git a/vscode-extension/src/commands/analyze.ts b/vscode-extension/src/commands/analyze.ts
index e01ea39..3dc93ec 100644
--- a/vscode-extension/src/commands/analyze.ts
+++ b/vscode-extension/src/commands/analyze.ts
@@ -14,7 +14,7 @@ export function setCachedAnalysis(analysis: RepoAnalysis | undefined): void {
 }
 
 export async function analyzeCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   await vscode.window.withProgress(
@@ -32,11 +32,18 @@ export async function analyzeCommand(): Promise<void> {
   );
 }
 
-export function getWorkspacePath(): string | undefined {
-  const folder = vscode.workspace.workspaceFolders?.[0];
-  if (!folder) {
+export async function pickWorkspacePath(): Promise<string | undefined> {
+  const folders = vscode.workspace.workspaceFolders;
+  if (!folders || folders.length === 0) {
     vscode.window.showWarningMessage("AgentRC: No workspace folder open.");
     return undefined;
   }
-  return folder.uri.fsPath;
+  if (folders.length === 1) {
+    return folders[0].uri.fsPath;
+  }
+  const picked = await vscode.window.showQuickPick(
+    folders.map((f) => ({ label: f.name, description: f.uri.fsPath, fsPath: f.uri.fsPath })),
+    { placeHolder: "Select workspace folder" }
+  );
+  return picked?.fsPath;
 }
diff --git a/vscode-extension/src/commands/batch.ts b/vscode-extension/src/commands/batch.ts
new file mode 100644
index 0000000..7ed64c6
--- /dev/null
+++ b/vscode-extension/src/commands/batch.ts
@@ -0,0 +1,78 @@
+import * as vscode from "vscode";
+import { generateCopilotInstructions, loadAgentrcConfig, safeWriteFile } from "../services.js";
+import { VscodeProgressReporter } from "../progress.js";
+import path from "node:path";
+
+export async function batchInstructionsCommand(): Promise<void> {
+  const folders = vscode.workspace.workspaceFolders;
+  if (!folders || folders.length === 0) {
+    vscode.window.showWarningMessage("AgentRC: No workspace folders open.");
+    return;
+  }
+  if (folders.length === 1) {
+    vscode.window.showInformationMessage(
+      "AgentRC: Only one workspace root — use 'Generate Instructions' instead."
+    );
+    return;
+  }
+
+  const model = vscode.workspace.getConfiguration("agentrc").get<string>("model");
+
+  await vscode.window.withProgress(
+    {
+      location: vscode.ProgressLocation.Notification,
+      title: "AgentRC: Generating instructions for all roots…",
+      cancellable: false
+    },
+    async (progress) => {
+      const reporter = new VscodeProgressReporter(progress);
+      let wrote = 0;
+      let skipped = 0;
+      let failed = 0;
+
+      for (const folder of folders) {
+        const workspacePath = folder.uri.fsPath;
+        const name = folder.name;
+        try {
+          let outputPath = path.join(workspacePath, ".github", "copilot-instructions.md");
+          try {
+            const config = await loadAgentrcConfig(workspacePath);
+            if (config?.strategy === "nested") {
+              outputPath = path.join(workspacePath, "AGENTS.md");
+            }
+          } catch {
+            // Non-fatal
+          }
+
+          reporter.update(`[${name}] Generating…`);
+          const content = await generateCopilotInstructions({
+            repoPath: workspacePath,
+            model
+          });
+
+          if (!content) {
+            skipped++;
+            continue;
+          }
+
+          const { wrote: didWrite, reason } = await safeWriteFile(outputPath, content, false);
+          if (didWrite) {
+            wrote++;
+          } else {
+            skipped++;
+            reporter.update(
+              `[${name}] Skipped: ${reason === "exists" ? "file already exists" : (reason ?? "unknown")}`
+            );
+          }
+        } catch (err) {
+          failed++;
+          reporter.update(`[${name}] Failed: ${err instanceof Error ? err.message : String(err)}`);
+        }
+      }
+
+      reporter.succeed(
+        `Done: ${wrote} generated, ${skipped} skipped, ${failed} failed (${folders.length} roots)`
+      );
+    }
+  );
+}
diff --git a/vscode-extension/src/commands/eval.ts b/vscode-extension/src/commands/eval.ts
index 7254fa1..938fae2 100644
--- a/vscode-extension/src/commands/eval.ts
+++ b/vscode-extension/src/commands/eval.ts
@@ -8,12 +8,12 @@ import {
   DEFAULT_MODEL
 } from "../services.js";
 import { VscodeProgressReporter } from "../progress.js";
-import { getWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
+import { pickWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
 import { createWebviewPanel } from "../webview.js";
 import fs from "node:fs";
 
 export async function evalCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   const configPath = path.join(workspacePath, "agentrc.eval.json");
@@ -68,7 +68,7 @@ export async function evalCommand(): Promise<void> {
 }
 
 export async function evalInitCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   const config = vscode.workspace.getConfiguration("agentrc");
diff --git a/vscode-extension/src/commands/generate.ts b/vscode-extension/src/commands/generate.ts
index f724225..dc9f38f 100644
--- a/vscode-extension/src/commands/generate.ts
+++ b/vscode-extension/src/commands/generate.ts
@@ -1,6 +1,6 @@
 import * as vscode from "vscode";
 import { generateConfigs, analyzeRepo } from "../services.js";
-import { getWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
+import { pickWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
 
 const GENERATE_OPTIONS = [
   { label: "MCP Config", value: "mcp", description: ".vscode/mcp.json" },
@@ -8,7 +8,7 @@ const GENERATE_OPTIONS = [
 ] as const;
 
 export async function generateCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   const picked = await vscode.window.showQuickPick(
diff --git a/vscode-extension/src/commands/init.ts b/vscode-extension/src/commands/init.ts
index 047aed2..0d365cf 100644
--- a/vscode-extension/src/commands/init.ts
+++ b/vscode-extension/src/commands/init.ts
@@ -7,10 +7,10 @@ import {
   safeWriteFile
 } from "../services.js";
 import { VscodeProgressReporter } from "../progress.js";
-import { getWorkspacePath, setCachedAnalysis } from "./analyze.js";
+import { pickWorkspacePath, setCachedAnalysis } from "./analyze.js";
 
 export async function initCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   const config = vscode.workspace.getConfiguration("agentrc");
@@ -30,7 +30,7 @@ export async function initCommand(): Promise<void> {
         const analysis = await analyzeRepo(workspacePath);
         setCachedAnalysis(analysis);
 
-        reporter.update("Generating Copilot instructions…");
+        reporter.update("Generating instructions…");
         const instructionsContent = await generateCopilotInstructions({
           repoPath: workspacePath,
           model,
diff --git a/vscode-extension/src/commands/instructions.ts b/vscode-extension/src/commands/instructions.ts
index 882a44a..9291041 100644
--- a/vscode-extension/src/commands/instructions.ts
+++ b/vscode-extension/src/commands/instructions.ts
@@ -1,5 +1,6 @@
 import * as vscode from "vscode";
 import path from "node:path";
+import fs from "node:fs";
 import {
   generateCopilotInstructions,
   generateAreaInstructions,
@@ -12,7 +13,7 @@ import {
   loadAgentrcConfig
 } from "../services.js";
 import { VscodeProgressReporter } from "../progress.js";
-import { getWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
+import { pickWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
 
 const FORMAT_OPTIONS = [
   {
@@ -43,7 +44,7 @@ const STRATEGY_OPTIONS = [
 ];
 
 export async function instructionsCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   const model = vscode.workspace.getConfiguration("agentrc").get<string>("model");
@@ -94,6 +95,40 @@ export async function instructionsCommand(): Promise<void> {
     if (claudePick) claudeMd = claudePick.value;
   }
 
+  // Write strategy choice back to agentrc.config.json
+  try {
+    const configPath = path.join(workspacePath, "agentrc.config.json");
+    let existing: Record<string, unknown> = {};
+    try {
+      const raw = await fs.promises.readFile(configPath, "utf-8");
+      const parsed: unknown = JSON.parse(raw);
+      if (parsed && typeof parsed === "object" && !Array.isArray(parsed)) {
+        existing = parsed as Record<string, unknown>;
+      }
+    } catch (err: unknown) {
+      if ((err as NodeJS.ErrnoException).code !== "ENOENT") {
+        throw err; // Malformed JSON — bubble to outer non-fatal catch
+      }
+    }
+    if (
+      existing.strategy !== strategy ||
+      existing.detailDir !== detailDir ||
+      existing.claudeMd !== claudeMd
+    ) {
+      existing.strategy = strategy;
+      if (strategy === "nested") {
+        existing.detailDir = detailDir;
+        existing.claudeMd = claudeMd;
+      } else {
+        delete existing.detailDir;
+        delete existing.claudeMd;
+      }
+      await safeWriteFile(configPath, JSON.stringify(existing, null, 2) + "\n", true);
+    }
+  } catch {
+    // Non-fatal — config write-back failure shouldn't block generation
+  }
+
   // Ensure analysis is available before starting progress
   let analysis = getCachedAnalysis();
   if (!analysis) {
@@ -111,7 +146,12 @@ export async function instructionsCommand(): Promise<void> {
   let selectedAreas: typeof analysis.areas = undefined;
   if (analysis.areas && analysis.areas.length > 0) {
     const picked = await vscode.window.showQuickPick(
-      analysis.areas.map((a) => ({ label: a.name, description: a.description, area: a })),
+      analysis.areas.map((a) => ({
+        label: a.name,
+        description: a.description,
+        detail: Array.isArray(a.applyTo) ? a.applyTo.join(", ") : a.applyTo,
+        area: a
+      })),
       { placeHolder: "Select areas for instructions (or Escape for root only)", canPickMany: true }
     );
     if (picked && picked.length > 0) {
diff --git a/vscode-extension/src/commands/pr.ts b/vscode-extension/src/commands/pr.ts
index 9bed3c3..4aaad71 100644
--- a/vscode-extension/src/commands/pr.ts
+++ b/vscode-extension/src/commands/pr.ts
@@ -6,11 +6,11 @@ import {
   isAgentrcFile
 } from "../services.js";
 import { getGitHubToken, getAzureDevOpsToken, detectPlatform, type PlatformInfo } from "../auth.js";
-import { getWorkspacePath } from "./analyze.js";
+import { pickWorkspacePath } from "./analyze.js";
 import { getGitRepository } from "../gitUtils.js";
 
 export async function prCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   const repository = getGitRepository(workspacePath);
diff --git a/vscode-extension/src/commands/readiness.ts b/vscode-extension/src/commands/readiness.ts
index 9995552..2266e09 100644
--- a/vscode-extension/src/commands/readiness.ts
+++ b/vscode-extension/src/commands/readiness.ts
@@ -1,18 +1,24 @@
 import * as vscode from "vscode";
-import { runReadinessReport, generateVisualReport, analyzeRepo } from "../services.js";
+import path from "node:path";
+import {
+  runReadinessReport,
+  generateVisualReport,
+  analyzeRepo,
+  loadAgentrcConfig
+} from "../services.js";
 import { VscodeProgressReporter } from "../progress.js";
-import { getWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
+import { pickWorkspacePath, getCachedAnalysis, setCachedAnalysis } from "./analyze.js";
 import { createWebviewPanel } from "../webview.js";
 import { readinessTreeProvider } from "../views/providers.js";
 
 export async function readinessCommand(): Promise<void> {
-  const workspacePath = getWorkspacePath();
+  const workspacePath = await pickWorkspacePath();
   if (!workspacePath) return;
 
   await vscode.window.withProgress(
     {
       location: vscode.ProgressLocation.Notification,
-      title: "AgentRC: Running readiness assessment…",
+      title: "AgentRC: Running readiness report…",
       cancellable: false
     },
     async (progress) => {
@@ -26,22 +32,33 @@ export async function readinessCommand(): Promise<void> {
           setCachedAnalysis(analysis);
         }
 
+        // Load policies from agentrc.config.json
+        let policies: string[] | undefined;
+        try {
+          const config = await loadAgentrcConfig(workspacePath);
+          policies = config?.policies;
+        } catch {
+          // Non-fatal — proceed without policies
+        }
+
         reporter.update("Evaluating readiness pillars…");
-        const report = await runReadinessReport({ repoPath: workspacePath });
+        const report = await runReadinessReport({ repoPath: workspacePath, policies });
 
         reporter.update("Generating report…");
-        const repoName = vscode.workspace.workspaceFolders?.[0]?.name ?? "Repository";
+        const repoName = path.basename(workspacePath);
         const html = generateVisualReport({
           reports: [{ repo: repoName, report }],
-          title: `${repoName} — AI Readiness`
+          title: `${repoName} — Readiness Report`
         });
 
-        createWebviewPanel("agentrc.readinessReport", "AI Readiness Report", html);
+        createWebviewPanel("agentrc.readinessReport", "Readiness Report", html);
         readinessTreeProvider.setReport(report);
-        reporter.succeed(`Readiness: Level ${report.achievedLevel} achieved.`);
+
+        const policyNote = policies?.length ? ` (${policies.length} policy source(s) applied)` : "";
+        reporter.succeed(`Readiness: Level ${report.achievedLevel} achieved.${policyNote}`);
       } catch (err) {
         vscode.window.showErrorMessage(
-          `AgentRC: Readiness assessment failed — ${err instanceof Error ? err.message : String(err)}`
+          `AgentRC: Readiness report failed — ${err instanceof Error ? err.message : String(err)}`
         );
       }
     }
diff --git a/vscode-extension/src/extension.ts b/vscode-extension/src/extension.ts
index b852802..a031b8c 100644
--- a/vscode-extension/src/extension.ts
+++ b/vscode-extension/src/extension.ts
@@ -6,6 +6,7 @@ import { readinessCommand } from "./commands/readiness.js";
 import { evalCommand, evalInitCommand } from "./commands/eval.js";
 import { initCommand } from "./commands/init.js";
 import { prCommand } from "./commands/pr.js";
+import { batchInstructionsCommand } from "./commands/batch.js";
 import { analysisTreeProvider, readinessTreeProvider } from "./views/providers.js";
 
 export function activate(context: vscode.ExtensionContext): void {
@@ -81,7 +82,8 @@ export function activate(context: vscode.ExtensionContext): void {
       updateStatusBar();
       vscode.commands.executeCommand("agentrc.analysis.focus");
     }),
-    vscode.commands.registerCommand("agentrc.pr", prCommand)
+    vscode.commands.registerCommand("agentrc.pr", prCommand),
+    vscode.commands.registerCommand("agentrc.batchInstructions", batchInstructionsCommand)
   );
 
   // Auto-analyze on activation if configured