wcag-a11y 0.3.6 → 0.4.0

package/README.md

@@ -56,7 +56,7 @@ Generating AI fixes for 7 violations...
 
 The **prompt** at the end of each fix is what you copy into Cursor, Copilot, or Claude. It includes the affected selectors, the WCAG rule, and exactly what needs to change — no rewriting needed.
 
-With `--report`, the full output is also saved to `a11y-report.md`.
+The full output is saved to `a11y-report.md` by default. Pass `--no-report` to skip.
 
 ---
 
@@ -69,17 +69,22 @@ npm install -g wcag-a11y
 ## Setup
 
 ```bash
-wcag-a11y init                       # Gemini (free, default)
-wcag-a11y init --provider openai     # OpenAI
-wcag-a11y init --provider ollama     # Local — no API key needed
+wcag-a11y init                            # Gemini (free, default)
+wcag-a11y init --provider openai          # OpenAI
+wcag-a11y init --provider anthropic       # Anthropic Claude
+wcag-a11y init --provider mistral         # Mistral
+wcag-a11y init --provider groq            # Groq (fast inference)
+wcag-a11y init --provider cohere          # Cohere
+wcag-a11y init --provider xai             # xAI Grok
+wcag-a11y init --provider deepseek        # DeepSeek
+wcag-a11y init --provider together        # Together AI (open-source models)
+wcag-a11y init --provider perplexity      # Perplexity
+wcag-a11y init --provider azure-openai    # Azure OpenAI
+wcag-a11y init --provider ollama          # Local — no API key needed
 ```
 
 Each command creates an `a11y.config.json` pre-wired for that provider. Fill in your API key, then scan.
 
-**Get a free Gemini key:** https://aistudio.google.com  
-**Get an OpenAI key:** https://platform.openai.com/api-keys  
-**Ollama (local):** install from https://ollama.com, then run `ollama serve`
-
 ---
 
 ## Commands
@@ -91,13 +96,13 @@ Scan a built-in page with 10 intentional violations. No dev server or config req
 ```bash
 wcag-a11y demo                  # violations + AI fixes (default, requires config)
 wcag-a11y demo --no-ai          # violations only, no AI — faster
-wcag-a11y demo --report         # + save a11y-report.md
+wcag-a11y demo --no-report      # skip saving a11y-report.md
 ```
 
 | Flag | Description |
 |---|---|
 | `--no-ai` | Skip AI fix generation — prints violations only, no prompts |
-| `-r, --report` | Save the full report to `a11y-report.md` in the current directory |
+| `--no-report` | Skip saving report to `a11y-report.md` (report is saved by default) |
 
 ---
 
@@ -113,7 +118,7 @@ wcag-a11y init --provider ollama     # Ollama (local)
 
 | Flag | Description |
 |---|---|
-| `--provider <name>` | Which provider to configure: `gemini` (default), `openai`, or `ollama`. Determines which fields are written to the config file. |
+| `--provider <name>` | Which provider to configure. Valid values: `gemini` (default), `openai`, `anthropic`, `mistral`, `groq`, `cohere`, `xai`, `deepseek`, `together`, `perplexity`, `azure-openai`, `ollama`. Determines which fields are written to the config file. |
 
 ---
 
@@ -124,8 +129,9 @@ Scan a running dev server for accessibility violations.
 ```bash
 wcag-a11y scan -u http://localhost:3000
 wcag-a11y scan -u http://localhost:3000 --pages / /about /contact
-wcag-a11y scan -u http://localhost:3000 --crawl --report
+wcag-a11y scan -u http://localhost:3000 --crawl
 wcag-a11y scan -u http://localhost:3000 --no-ai --ci
+wcag-a11y scan -u http://localhost:3000 --terminal --fast-mode
 ```
 
 | Flag | Default | Description |
@@ -133,12 +139,14 @@ wcag-a11y scan -u http://localhost:3000 --no-ai --ci
 | `-u, --url <url>` | required | Base URL of your running dev server |
 | `-p, --pages <pages...>` | `/` | One or more paths to scan. Separate with spaces: `--pages / /about /contact` |
 | `-c, --crawl` | off | Follow same-origin links and scan all reachable pages automatically |
-| `-r, --report` | off | Save the full scan output to `a11y-report.md` |
+| `--no-report` | on | Skip saving scan output to `a11y-report.md` (report is saved by default) |
 | `--no-ai` | on | Skip AI fix generation — scan runs faster and prints violations only |
-| `--no-explain` | off | Print only the ready-to-paste prompt for each fix, without the AI explanation |
+| `--no-explain` | on | Print only the ready-to-paste prompt for each fix, without the AI explanation |
+| `--terminal` | off | Print violations summary and AI fix prompts to terminal |
+| `--fast-mode` | off | Output only AI fix prompts — no summaries, explanations, or progress messages |
 | `--group <strategy>` | `rule` | `rule` (default) groups all violations of the same type into one fix prompt. `none` produces a separate prompt per element. Use `none` when violations of the same rule need different fixes |
 | `--ci` | off | Exit with code `1` if any violations are found. Use this to fail a CI pipeline |
-| `--provider <name>` | from config | Override the AI provider for this run: `gemini`, `openai`, or `ollama`. Does not modify the config file |
+| `--provider <name>` | from config | Override the AI provider for this run. See [AI Providers](#ai-providers) for valid names. Does not modify the config file |
 
 ---
 
@@ -195,33 +203,46 @@ src/components/Navbar.jsx — 2 violation(s)
 | `-c, --crawl` | off | Auto-discover pages by following same-origin links |
 | `--from-report [path]` | `a11y-report.md` | Load violations from an existing report instead of scanning. Useful when you already ran `scan --report` and just want to apply fixes |
 | `--apply` | off | Write patched files to disk (dry-run without this flag) |
-| `--provider <name>` | from config | Override AI provider for this run: `gemini`, `openai`, or `ollama` |
+| `--provider <name>` | from config | Override AI provider for this run. See [AI Providers](#ai-providers) for valid names |
 
 > **Tip:** Always run without `--apply` first to review the diff. The dry-run is safe — nothing is written to disk.
 
 **Common workflow:** run `scan --report` to generate a report for review, then run `fix --from-report --apply` to patch the files — no second browser crawl needed.
 
 ```bash
-wcag-a11y scan -u http://localhost:3000 --report   # review a11y-report.md
-wcag-a11y fix --from-report --apply                 # patch files from that report
+wcag-a11y scan -u http://localhost:3000            # generates a11y-report.md automatically
+wcag-a11y fix --from-report --apply                # patch files from that report
 ```
 
 ---
 
 ## AI Providers
 
-| Provider | Model | Cost | API Key |
-|---|---|---|---|
-| `gemini` (default) | `gemini-2.5-flash` | Free tier | [aistudio.google.com](https://aistudio.google.com) |
-| `openai` | `gpt-4o-mini` | Pay-per-use | [platform.openai.com](https://platform.openai.com/api-keys) |
-| `ollama` | `llama3` | Free (local) | None — run `ollama serve` |
+12 providers are supported. Set your provider in `a11y.config.json` or override per-run with `--provider <name>`.
 
-Set your provider in `a11y.config.json` or override it per-run with `--provider`. If the AI response is unparseable, the tool generates a fix prompt directly from the violation data so you always get something actionable.
+| Provider | `--provider` name | Default model | API key source |
+|---|---|---|---|
+| Google Gemini | `gemini` *(default)* | `gemini-2.5-flash` | [aistudio.google.com](https://aistudio.google.com) — free tier |
+| OpenAI | `openai` | `gpt-4o-mini` | [platform.openai.com/api-keys](https://platform.openai.com/api-keys) |
+| Anthropic | `anthropic` | `claude-sonnet-4-6` | [console.anthropic.com](https://console.anthropic.com) |
+| Mistral | `mistral` | `mistral-large-latest` | [console.mistral.ai](https://console.mistral.ai) |
+| Groq | `groq` | `llama-3.3-70b-versatile` | [console.groq.com](https://console.groq.com) |
+| Cohere | `cohere` | `command-r-plus` | [dashboard.cohere.com](https://dashboard.cohere.com) |
+| xAI | `xai` | `grok-2` | [console.x.ai](https://console.x.ai) |
+| DeepSeek | `deepseek` | `deepseek-chat` | [platform.deepseek.com](https://platform.deepseek.com) |
+| Together AI | `together` | `meta-llama/Llama-3-70b-chat-hf` | [api.together.xyz](https://api.together.xyz) |
+| Perplexity | `perplexity` | `llama-3.1-sonar-large-128k-online` | [perplexity.ai/settings/api](https://www.perplexity.ai/settings/api) |
+| Azure OpenAI | `azure-openai` | *(your deployment)* | [portal.azure.com](https://portal.azure.com) |
+| Ollama | `ollama` | `llama3` | None — run `ollama serve` locally |
+
+All models are configurable. Change the model field in `a11y.config.json` to use any model your API key has access to. If the AI response is unparseable, the tool generates a fix prompt directly from the violation data so you always get something actionable.
 
 ---
 
 ## Config (`a11y.config.json`)
 
+Only the fields for your active `provider` are required. This file is gitignored by default.
+
 ```json
 {
   "provider": "gemini",
@@ -232,13 +253,40 @@ Set your provider in `a11y.config.json` or override it per-run with `--provider`
   "openaiApiKey": "YOUR_OPENAI_API_KEY",
   "openaiModel": "gpt-4o-mini",
 
+  "anthropicApiKey": "YOUR_ANTHROPIC_API_KEY",
+  "anthropicModel": "claude-sonnet-4-6",
+
+  "mistralApiKey": "YOUR_MISTRAL_API_KEY",
+  "mistralModel": "mistral-large-latest",
+
+  "groqApiKey": "YOUR_GROQ_API_KEY",
+  "groqModel": "llama-3.3-70b-versatile",
+
+  "cohereApiKey": "YOUR_COHERE_API_KEY",
+  "cohereModel": "command-r-plus",
+
+  "xaiApiKey": "YOUR_XAI_API_KEY",
+  "xaiModel": "grok-2",
+
+  "deepseekApiKey": "YOUR_DEEPSEEK_API_KEY",
+  "deepseekModel": "deepseek-chat",
+
+  "togetherApiKey": "YOUR_TOGETHER_API_KEY",
+  "togetherModel": "meta-llama/Llama-3-70b-chat-hf",
+
+  "perplexityApiKey": "YOUR_PERPLEXITY_API_KEY",
+  "perplexityModel": "llama-3.1-sonar-large-128k-online",
+
+  "azureOpenaiApiKey": "YOUR_AZURE_KEY",
+  "azureOpenaiEndpoint": "https://YOUR_RESOURCE.openai.azure.com",
+  "azureOpenaiDeployment": "YOUR_DEPLOYMENT_NAME",
+  "azureOpenaiApiVersion": "2024-10-01-preview",
+
   "ollamaBaseUrl": "http://localhost:11434",
   "ollamaModel": "llama3"
 }
 ```
 
-Only the fields for your active provider are required. This file is gitignored by default.
-
 ---
 
 ## What it checks

package/dist/ai/anthropic.js

@@ -0,0 +1,59 @@
+import { buildPrompt } from './prompt.js';
+import { buildPatchPrompt } from './patch-prompt.js';
+import { groupViolations } from './group.js';
+import { BaseAIProvider } from './base.js';
+export class AnthropicProvider extends BaseAIProvider {
+    apiKey;
+    model;
+    constructor(apiKey, model = 'claude-sonnet-4-6') {
+        super();
+        this.apiKey = apiKey;
+        this.model = model;
+    }
+    get headers() {
+        return {
+            'Content-Type': 'application/json',
+            'x-api-key': this.apiKey,
+            'anthropic-version': '2023-06-01',
+        };
+    }
+    async generateFixes(violations, strategy = 'rule', framework) {
+        if (violations.length === 0)
+            return [];
+        const groups = groupViolations(violations, strategy);
+        const prompt = buildPrompt(groups, framework);
+        const response = await fetch('https://api.anthropic.com/v1/messages', {
+            method: 'POST',
+            headers: this.headers,
+            body: JSON.stringify({
+                model: this.model,
+                max_tokens: 8192,
+                temperature: 0.2,
+                messages: [{ role: 'user', content: prompt }],
+            }),
+        });
+        if (!response.ok) {
+            throw new Error(`Anthropic API error: ${response.status} ${await response.text()}`);
+        }
+        const data = await response.json();
+        const text = data.content?.[0]?.text ?? '[]';
+        return this.parse(text, groups, framework);
+    }
+    async generateFilePatch(fileContent, violations, filePath, framework) {
+        const prompt = buildPatchPrompt(fileContent, violations, filePath, framework);
+        const response = await fetch('https://api.anthropic.com/v1/messages', {
+            method: 'POST',
+            headers: this.headers,
+            body: JSON.stringify({
+                model: this.model,
+                max_tokens: 16384,
+                temperature: 0.1,
+                messages: [{ role: 'user', content: prompt }],
+            }),
+        });
+        if (!response.ok)
+            throw new Error(`Anthropic API error: ${response.status} ${await response.text()}`);
+        const data = await response.json();
+        return data.content?.[0]?.text ?? '';
+    }
+}

package/dist/ai/azure-openai.js

@@ -0,0 +1,17 @@
+import { OpenAICompatProvider } from './openai-compat.js';
+export class AzureOpenAIProvider extends OpenAICompatProvider {
+    apiVersion;
+    constructor(endpoint, apiKey, deployment, apiVersion = '2024-10-01-preview') {
+        super(`${endpoint}/openai/deployments/${deployment}`, apiKey, deployment, 'Azure OpenAI');
+        this.apiVersion = apiVersion;
+    }
+    buildUrl(path) {
+        return `${this.baseUrl}${path}?api-version=${this.apiVersion}`;
+    }
+    buildHeaders() {
+        return {
+            'Content-Type': 'application/json',
+            'api-key': this.apiKey,
+        };
+    }
+}

package/dist/ai/base.js

@@ -0,0 +1,39 @@
+import { fallbackExplanation } from './fallback-explanation.js';
+export class BaseAIProvider {
+    parse(text, groups, framework) {
+        try {
+            const jsonMatch = text.match(/\[[\s\S]*\]/);
+            const fixes = jsonMatch ? JSON.parse(jsonMatch[0]) : [];
+            return groups.map((g) => {
+                const found = fixes.find((f) => f.ruleId === g.ruleId);
+                return found
+                    ? { ...found, selectors: g.selectors, instanceCount: g.count }
+                    : this.fallbackFix(g, framework);
+            });
+        }
+        catch {
+            return groups.map((g) => this.fallbackFix(g, framework));
+        }
+    }
+    fallback(groups, framework) {
+        return groups.map((g) => this.fallbackFix(g, framework));
+    }
+    fallbackFix(g, framework) {
+        const v = g.representative;
+        const selectorList = g.selectors.map((s) => `- ${s}`).join('\n');
+        const explanation = fallbackExplanation(g.ruleId, g.description, g.wcag, g.level);
+        const fwNote = framework ? `This project uses ${framework}. ` : '';
+        const prompt = g.count > 1
+            ? `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected elements (${g.count} instances):\n${selectorList}\n\nRepresentative HTML:\n\`${v.html.slice(0, 300)}\`\n\nApply the fix to all ${g.count} instances in the codebase to comply with WCAG 2.1 SC ${g.wcag}.`
+            : `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected element:\n- Selector: \`${g.selectors[0]}\`\n- HTML: \`${v.html.slice(0, 300)}\`\n\nApply the fix to comply with WCAG 2.1 SC ${g.wcag}.`;
+        return {
+            ruleId: g.ruleId,
+            selectors: g.selectors,
+            instanceCount: g.count,
+            explanation,
+            fixedCode: v.html,
+            wcagReference: `WCAG 2.1 SC ${g.wcag}`,
+            optimalPrompt: prompt,
+        };
+    }
+}

package/dist/ai/cohere.js

@@ -0,0 +1,58 @@
+import { buildPrompt } from './prompt.js';
+import { buildPatchPrompt } from './patch-prompt.js';
+import { groupViolations } from './group.js';
+import { BaseAIProvider } from './base.js';
+export class CohereProvider extends BaseAIProvider {
+    apiKey;
+    model;
+    constructor(apiKey, model = 'command-r-plus') {
+        super();
+        this.apiKey = apiKey;
+        this.model = model;
+    }
+    get headers() {
+        return {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${this.apiKey}`,
+        };
+    }
+    async generateFixes(violations, strategy = 'rule', framework) {
+        if (violations.length === 0)
+            return [];
+        const groups = groupViolations(violations, strategy);
+        const prompt = buildPrompt(groups, framework);
+        const response = await fetch('https://api.cohere.com/v2/chat', {
+            method: 'POST',
+            headers: this.headers,
+            body: JSON.stringify({
+                model: this.model,
+                messages: [{ role: 'user', content: prompt }],
+                max_tokens: 8192,
+                temperature: 0.2,
+            }),
+        });
+        if (!response.ok) {
+            throw new Error(`Cohere API error: ${response.status} ${await response.text()}`);
+        }
+        const data = await response.json();
+        const text = data.message?.content?.[0]?.text ?? '[]';
+        return this.parse(text, groups, framework);
+    }
+    async generateFilePatch(fileContent, violations, filePath, framework) {
+        const prompt = buildPatchPrompt(fileContent, violations, filePath, framework);
+        const response = await fetch('https://api.cohere.com/v2/chat', {
+            method: 'POST',
+            headers: this.headers,
+            body: JSON.stringify({
+                model: this.model,
+                messages: [{ role: 'user', content: prompt }],
+                max_tokens: 16384,
+                temperature: 0.1,
+            }),
+        });
+        if (!response.ok)
+            throw new Error(`Cohere API error: ${response.status} ${await response.text()}`);
+        const data = await response.json();
+        return data.message?.content?.[0]?.text ?? '';
+    }
+}

package/dist/ai/deepseek.js

@@ -0,0 +1,6 @@
+import { OpenAICompatProvider } from './openai-compat.js';
+export class DeepSeekProvider extends OpenAICompatProvider {
+    constructor(apiKey, model = 'deepseek-chat') {
+        super('https://api.deepseek.com/v1', apiKey, model, 'DeepSeek');
+    }
+}

package/dist/ai/gemini.js

@@ -1,21 +1,24 @@
 import { buildPrompt } from './prompt.js';
 import { buildPatchPrompt } from './patch-prompt.js';
 import { groupViolations } from './group.js';
-import { fallbackExplanation } from './fallback-explanation.js';
-export class GeminiProvider {
+import { BaseAIProvider } from './base.js';
+export class GeminiProvider extends BaseAIProvider {
     apiKey;
     model;
     constructor(apiKey, model = 'gemini-2.5-flash') {
+        super();
         this.apiKey = apiKey;
         this.model = model;
     }
+    buildUrl(model) {
+        return `https://generativelanguage.googleapis.com/v1beta/models/${model}:generateContent?key=${this.apiKey}`;
+    }
     async generateFixes(violations, strategy = 'rule', framework) {
         if (violations.length === 0)
             return [];
         const groups = groupViolations(violations, strategy);
         const prompt = buildPrompt(groups, framework);
-        const url = `https://generativelanguage.googleapis.com/v1beta/models/${this.model}:generateContent?key=${this.apiKey}`;
-        const response = await fetch(url, {
+        const response = await fetch(this.buildUrl(this.model), {
             method: 'POST',
             headers: { 'Content-Type': 'application/json' },
             body: JSON.stringify({
@@ -32,8 +35,7 @@ export class GeminiProvider {
     }
     async generateFilePatch(fileContent, violations, filePath, framework) {
         const prompt = buildPatchPrompt(fileContent, violations, filePath, framework);
-        const url = `https://generativelanguage.googleapis.com/v1beta/models/${this.model}:generateContent?key=${this.apiKey}`;
-        const response = await fetch(url, {
+        const response = await fetch(this.buildUrl(this.model), {
             method: 'POST',
             headers: { 'Content-Type': 'application/json' },
             body: JSON.stringify({
@@ -46,37 +48,4 @@ export class GeminiProvider {
         const data = await response.json();
         return data.candidates?.[0]?.content?.parts?.[0]?.text ?? '';
     }
-    parse(text, groups, framework) {
-        try {
-            const jsonMatch = text.match(/\[[\s\S]*\]/);
-            const fixes = jsonMatch ? JSON.parse(jsonMatch[0]) : [];
-            return groups.map((g) => {
-                const found = fixes.find((f) => f.ruleId === g.ruleId);
-                return found
-                    ? { ...found, selectors: g.selectors, instanceCount: g.count }
-                    : this.fallbackFix(g, framework);
-            });
-        }
-        catch {
-            return groups.map((g) => this.fallbackFix(g, framework));
-        }
-    }
-    fallbackFix(g, framework) {
-        const v = g.representative;
-        const selectorList = g.selectors.map((s) => `- ${s}`).join('\n');
-        const explanation = fallbackExplanation(g.ruleId, g.description, g.wcag, g.level);
-        const fwNote = framework ? `This project uses ${framework}. ` : '';
-        const prompt = g.count > 1
-            ? `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected elements (${g.count} instances):\n${selectorList}\n\nRepresentative HTML:\n\`${v.html.slice(0, 300)}\`\n\nApply the fix to all ${g.count} instances in the codebase to comply with WCAG 2.1 SC ${g.wcag}.`
-            : `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected element:\n- Selector: \`${g.selectors[0]}\`\n- HTML: \`${v.html.slice(0, 300)}\`\n\nApply the fix to comply with WCAG 2.1 SC ${g.wcag}.`;
-        return {
-            ruleId: g.ruleId,
-            selectors: g.selectors,
-            instanceCount: g.count,
-            explanation,
-            fixedCode: v.html,
-            wcagReference: `WCAG 2.1 SC ${g.wcag}`,
-            optimalPrompt: prompt,
-        };
-    }
 }

package/dist/ai/groq.js

@@ -0,0 +1,6 @@
+import { OpenAICompatProvider } from './openai-compat.js';
+export class GroqProvider extends OpenAICompatProvider {
+    constructor(apiKey, model = 'llama-3.3-70b-versatile') {
+        super('https://api.groq.com/openai/v1', apiKey, model, 'Groq');
+    }
+}

package/dist/ai/index.js

@@ -1,20 +1,84 @@
 import { GeminiProvider } from './gemini.js';
 import { OllamaProvider } from './ollama.js';
 import { OpenAIProvider } from './openai.js';
+import { AnthropicProvider } from './anthropic.js';
+import { MistralProvider } from './mistral.js';
+import { GroqProvider } from './groq.js';
+import { CohereProvider } from './cohere.js';
+import { XAIProvider } from './xai.js';
+import { DeepSeekProvider } from './deepseek.js';
+import { TogetherProvider } from './together.js';
+import { PerplexityProvider } from './perplexity.js';
+import { AzureOpenAIProvider } from './azure-openai.js';
 export function createAIProvider(config) {
-    if (config.provider === 'ollama') {
-        return new OllamaProvider(config.ollamaBaseUrl, config.ollamaModel);
+    switch (config.provider) {
+        case 'ollama':
+            return new OllamaProvider(config.ollamaBaseUrl, config.ollamaModel);
+        case 'openai':
+            if (!config.openaiApiKey) {
+                console.error('No OpenAI API key found. Add "openaiApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new OpenAIProvider(config.openaiApiKey, config.openaiModel);
+        case 'anthropic':
+            if (!config.anthropicApiKey) {
+                console.error('No Anthropic API key found. Add "anthropicApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new AnthropicProvider(config.anthropicApiKey, config.anthropicModel);
+        case 'mistral':
+            if (!config.mistralApiKey) {
+                console.error('No Mistral API key found. Add "mistralApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new MistralProvider(config.mistralApiKey, config.mistralModel);
+        case 'groq':
+            if (!config.groqApiKey) {
+                console.error('No Groq API key found. Add "groqApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new GroqProvider(config.groqApiKey, config.groqModel);
+        case 'cohere':
+            if (!config.cohereApiKey) {
+                console.error('No Cohere API key found. Add "cohereApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new CohereProvider(config.cohereApiKey, config.cohereModel);
+        case 'xai':
+            if (!config.xaiApiKey) {
+                console.error('No xAI API key found. Add "xaiApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new XAIProvider(config.xaiApiKey, config.xaiModel);
+        case 'deepseek':
+            if (!config.deepseekApiKey) {
+                console.error('No DeepSeek API key found. Add "deepseekApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new DeepSeekProvider(config.deepseekApiKey, config.deepseekModel);
+        case 'together':
+            if (!config.togetherApiKey) {
+                console.error('No Together AI API key found. Add "togetherApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new TogetherProvider(config.togetherApiKey, config.togetherModel);
+        case 'perplexity':
+            if (!config.perplexityApiKey) {
+                console.error('No Perplexity API key found. Add "perplexityApiKey" to a11y.config.json.');
+                process.exit(1);
+            }
+            return new PerplexityProvider(config.perplexityApiKey, config.perplexityModel);
+        case 'azure-openai':
+            if (!config.azureOpenaiApiKey || !config.azureOpenaiEndpoint || !config.azureOpenaiDeployment) {
+                console.error('Azure OpenAI requires "azureOpenaiApiKey", "azureOpenaiEndpoint", and "azureOpenaiDeployment" in a11y.config.json.');
+                process.exit(1);
+            }
+            return new AzureOpenAIProvider(config.azureOpenaiEndpoint, config.azureOpenaiApiKey, config.azureOpenaiDeployment, config.azureOpenaiApiVersion);
+        default:
+            if (!config.apiKey) {
+                console.error('No API key found. Run: wcag-a11y init');
+                process.exit(1);
+            }
+            return new GeminiProvider(config.apiKey, config.model);
     }
-    if (config.provider === 'openai') {
-        if (!config.openaiApiKey) {
-            console.error('No OpenAI API key found. Add "openaiApiKey" to a11y.config.json.');
-            process.exit(1);
-        }
-        return new OpenAIProvider(config.openaiApiKey, config.openaiModel);
-    }
-    if (!config.apiKey) {
-        console.error('No API key found. Run: wcag-a11y init');
-        process.exit(1);
-    }
-    return new GeminiProvider(config.apiKey, config.model);
 }

package/dist/ai/mistral.js

@@ -0,0 +1,6 @@
+import { OpenAICompatProvider } from './openai-compat.js';
+export class MistralProvider extends OpenAICompatProvider {
+    constructor(apiKey, model = 'mistral-large-latest') {
+        super('https://api.mistral.ai/v1', apiKey, model, 'Mistral');
+    }
+}

package/dist/ai/ollama.js

@@ -1,11 +1,12 @@
 import { buildPrompt } from './prompt.js';
 import { buildPatchPrompt } from './patch-prompt.js';
 import { groupViolations } from './group.js';
-import { fallbackExplanation } from './fallback-explanation.js';
-export class OllamaProvider {
+import { BaseAIProvider } from './base.js';
+export class OllamaProvider extends BaseAIProvider {
     baseUrl;
     model;
     constructor(baseUrl = 'http://localhost:11434', model = 'llama3') {
+        super();
         this.baseUrl = baseUrl;
         this.model = model;
     }
@@ -52,25 +53,4 @@ export class OllamaProvider {
         const data = await response.json();
         return data.response ?? '';
     }
-    fallback(groups, framework) {
-        return groups.map((g) => this.fallbackFix(g, framework));
-    }
-    fallbackFix(g, framework) {
-        const v = g.representative;
-        const selectorList = g.selectors.map((s) => `- ${s}`).join('\n');
-        const explanation = fallbackExplanation(g.ruleId, g.description, g.wcag, g.level);
-        const fwNote = framework ? `This project uses ${framework}. ` : '';
-        const prompt = g.count > 1
-            ? `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected elements (${g.count} instances):\n${selectorList}\n\nRepresentative HTML:\n\`${v.html.slice(0, 300)}\`\n\nApply the fix to all ${g.count} instances in the codebase to comply with WCAG 2.1 SC ${g.wcag}.`
-            : `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected element:\n- Selector: \`${g.selectors[0]}\`\n- HTML: \`${v.html.slice(0, 300)}\`\n\nApply the fix to comply with WCAG 2.1 SC ${g.wcag}.`;
-        return {
-            ruleId: g.ruleId,
-            selectors: g.selectors,
-            instanceCount: g.count,
-            explanation,
-            fixedCode: v.html,
-            wcagReference: `WCAG 2.1 SC ${g.wcag}`,
-            optimalPrompt: prompt,
-        };
-    }
 }

package/dist/ai/openai-compat.js

@@ -0,0 +1,65 @@
+import { buildPrompt } from './prompt.js';
+import { buildPatchPrompt } from './patch-prompt.js';
+import { groupViolations } from './group.js';
+import { BaseAIProvider } from './base.js';
+export class OpenAICompatProvider extends BaseAIProvider {
+    baseUrl;
+    apiKey;
+    model;
+    providerName;
+    constructor(baseUrl, apiKey, model, providerName = 'API') {
+        super();
+        this.baseUrl = baseUrl;
+        this.apiKey = apiKey;
+        this.model = model;
+        this.providerName = providerName;
+    }
+    buildUrl(path) {
+        return `${this.baseUrl}${path}`;
+    }
+    buildHeaders() {
+        return {
+            'Content-Type': 'application/json',
+            'Authorization': `Bearer ${this.apiKey}`,
+        };
+    }
+    async generateFixes(violations, strategy = 'rule', framework) {
+        if (violations.length === 0)
+            return [];
+        const groups = groupViolations(violations, strategy);
+        const prompt = buildPrompt(groups, framework);
+        const response = await fetch(this.buildUrl('/chat/completions'), {
+            method: 'POST',
+            headers: this.buildHeaders(),
+            body: JSON.stringify({
+                model: this.model,
+                messages: [{ role: 'user', content: prompt }],
+                temperature: 0.2,
+                max_tokens: 8192,
+            }),
+        });
+        if (!response.ok) {
+            throw new Error(`${this.providerName} API error: ${response.status} ${await response.text()}`);
+        }
+        const data = await response.json();
+        const text = data.choices?.[0]?.message?.content ?? '[]';
+        return this.parse(text, groups, framework);
+    }
+    async generateFilePatch(fileContent, violations, filePath, framework) {
+        const prompt = buildPatchPrompt(fileContent, violations, filePath, framework);
+        const response = await fetch(this.buildUrl('/chat/completions'), {
+            method: 'POST',
+            headers: this.buildHeaders(),
+            body: JSON.stringify({
+                model: this.model,
+                messages: [{ role: 'user', content: prompt }],
+                temperature: 0.1,
+                max_tokens: 16384,
+            }),
+        });
+        if (!response.ok)
+            throw new Error(`${this.providerName} API error: ${response.status} ${await response.text()}`);
+        const data = await response.json();
+        return data.choices?.[0]?.message?.content ?? '';
+    }
+}

package/dist/ai/openai.js

@@ -1,87 +1,6 @@
-import { buildPrompt } from './prompt.js';
-import { buildPatchPrompt } from './patch-prompt.js';
-import { groupViolations } from './group.js';
-import { fallbackExplanation } from './fallback-explanation.js';
-export class OpenAIProvider {
-    apiKey;
-    model;
+import { OpenAICompatProvider } from './openai-compat.js';
+export class OpenAIProvider extends OpenAICompatProvider {
     constructor(apiKey, model = 'gpt-4o-mini') {
-        this.apiKey = apiKey;
-        this.model = model;
-    }
-    async generateFixes(violations, strategy = 'rule', framework) {
-        if (violations.length === 0)
-            return [];
-        const groups = groupViolations(violations, strategy);
-        const prompt = buildPrompt(groups, framework);
-        const response = await fetch('https://api.openai.com/v1/chat/completions', {
-            method: 'POST',
-            headers: {
-                'Content-Type': 'application/json',
-                'Authorization': `Bearer ${this.apiKey}`,
-            },
-            body: JSON.stringify({
-                model: this.model,
-                messages: [{ role: 'user', content: prompt }],
-                temperature: 0.2,
-                max_tokens: 8192,
-            }),
-        });
-        if (!response.ok) {
-            throw new Error(`OpenAI API error: ${response.status} ${await response.text()}`);
-        }
-        const data = await response.json();
-        const text = data.choices?.[0]?.message?.content ?? '[]';
-        return this.parse(text, groups, framework);
-    }
-    async generateFilePatch(fileContent, violations, filePath, framework) {
-        const prompt = buildPatchPrompt(fileContent, violations, filePath, framework);
-        const response = await fetch('https://api.openai.com/v1/chat/completions', {
-            method: 'POST',
-            headers: { 'Content-Type': 'application/json', Authorization: `Bearer ${this.apiKey}` },
-            body: JSON.stringify({
-                model: this.model,
-                messages: [{ role: 'user', content: prompt }],
-                temperature: 0.1,
-                max_tokens: 16384,
-            }),
-        });
-        if (!response.ok)
-            throw new Error(`OpenAI API error: ${response.status} ${await response.text()}`);
-        const data = await response.json();
-        return data.choices?.[0]?.message?.content ?? '';
-    }
-    parse(text, groups, framework) {
-        try {
-            const jsonMatch = text.match(/\[[\s\S]*\]/);
-            const fixes = jsonMatch ? JSON.parse(jsonMatch[0]) : [];
-            return groups.map((g) => {
-                const found = fixes.find((f) => f.ruleId === g.ruleId);
-                return found
-                    ? { ...found, selectors: g.selectors, instanceCount: g.count }
-                    : this.fallbackFix(g, framework);
-            });
-        }
-        catch {
-            return groups.map((g) => this.fallbackFix(g, framework));
-        }
-    }
-    fallbackFix(g, framework) {
-        const v = g.representative;
-        const selectorList = g.selectors.map((s) => `- ${s}`).join('\n');
-        const explanation = fallbackExplanation(g.ruleId, g.description, g.wcag, g.level);
-        const fwNote = framework ? `This project uses ${framework}. ` : '';
-        const prompt = g.count > 1
-            ? `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected elements (${g.count} instances):\n${selectorList}\n\nRepresentative HTML:\n\`${v.html.slice(0, 300)}\`\n\nApply the fix to all ${g.count} instances in the codebase to comply with WCAG 2.1 SC ${g.wcag}.`
-            : `${fwNote}Fix WCAG 2.1 SC ${g.wcag} (Level ${g.level}) — ${g.description}\n\nAffected element:\n- Selector: \`${g.selectors[0]}\`\n- HTML: \`${v.html.slice(0, 300)}\`\n\nApply the fix to comply with WCAG 2.1 SC ${g.wcag}.`;
-        return {
-            ruleId: g.ruleId,
-            selectors: g.selectors,
-            instanceCount: g.count,
-            explanation,
-            fixedCode: v.html,
-            wcagReference: `WCAG 2.1 SC ${g.wcag}`,
-            optimalPrompt: prompt,
-        };
+        super('https://api.openai.com/v1', apiKey, model, 'OpenAI');
     }
 }

package/dist/ai/perplexity.js

@@ -0,0 +1,6 @@
+import { OpenAICompatProvider } from './openai-compat.js';
+export class PerplexityProvider extends OpenAICompatProvider {
+    constructor(apiKey, model = 'llama-3.1-sonar-large-128k-online') {
+        super('https://api.perplexity.ai', apiKey, model, 'Perplexity');
+    }
+}

package/dist/ai/together.js

@@ -0,0 +1,6 @@
+import { OpenAICompatProvider } from './openai-compat.js';
+export class TogetherProvider extends OpenAICompatProvider {
+    constructor(apiKey, model = 'meta-llama/Llama-3-70b-chat-hf') {
+        super('https://api.together.xyz/v1', apiKey, model, 'Together AI');
+    }
+}

package/dist/ai/xai.js

@@ -0,0 +1,6 @@
+import { OpenAICompatProvider } from './openai-compat.js';
+export class XAIProvider extends OpenAICompatProvider {
+    constructor(apiKey, model = 'grok-2') {
+        super('https://api.x.ai/v1', apiKey, model, 'xAI');
+    }
+}

package/dist/cli.js

@@ -13,11 +13,11 @@ const program = new Command();
 program
     .name('wcag-a11y')
     .description('WCAG 2.1/2.2 accessibility auditor with AI-powered fixes')
-    .version('0.3.6');
+    .version('0.4.0');
 program
     .command('init')
     .description('Create a11y.config.json in the current directory')
-    .option('--provider <name>', 'AI provider to configure (gemini|openai|ollama)', 'gemini')
+    .option('--provider <name>', 'AI provider to configure (gemini|openai|ollama|anthropic|mistral|groq|cohere|xai|deepseek|together|perplexity|azure-openai)', 'gemini')
     .action((opts) => {
     initConfig(opts.provider);
 });
@@ -34,7 +34,7 @@ program
     .option('--fast-mode', 'Output only AI fix prompts — no summaries or explanations', false)
     .option('--group <strategy>', 'Group violations by rule or show individually (rule|none)', 'rule')
     .option('--ci', 'Exit with code 1 if any violations are found (for CI/CD pipelines)', false)
-    .option('--provider <name>', 'Override the AI provider from config (gemini|openai|ollama)')
+    .option('--provider <name>', 'Override the AI provider from config (gemini|openai|ollama|anthropic|mistral|groq|cohere|xai|deepseek|together|perplexity|azure-openai)')
     .action(async (opts) => {
     try {
         console.log(`\nScanning ${opts.url}...`);
@@ -82,7 +82,7 @@ program
     .option('-c, --crawl', 'Auto-discover pages by following same-origin links', false)
     .option('--from-report [path]', 'Use an existing report instead of scanning (default: a11y-report.md)')
     .option('--apply', 'Write fixes to source files (default: dry-run, shows diff only)', false)
-    .option('--provider <name>', 'Override the AI provider from config (gemini|openai|ollama)')
+    .option('--provider <name>', 'Override the AI provider from config (gemini|openai|ollama|anthropic|mistral|groq|cohere|xai|deepseek|together|perplexity|azure-openai)')
     .action(async (opts) => {
     if (!opts.url && !opts.fromReport) {
         console.error('\nError: provide --url <url> to scan, or --from-report [path] to load an existing report.');

package/dist/config.js

@@ -7,6 +7,15 @@ const DEFAULTS = {
     ollamaBaseUrl: 'http://localhost:11434',
     ollamaModel: 'llama3',
     openaiModel: 'gpt-4o-mini',
+    anthropicModel: 'claude-sonnet-4-6',
+    mistralModel: 'mistral-large-latest',
+    groqModel: 'llama-3.3-70b-versatile',
+    cohereModel: 'command-r-plus',
+    xaiModel: 'grok-2',
+    deepseekModel: 'deepseek-chat',
+    togetherModel: 'meta-llama/Llama-3-70b-chat-hf',
+    perplexityModel: 'llama-3.1-sonar-large-128k-online',
+    azureOpenaiApiVersion: '2024-10-01-preview',
 };
 export function loadConfig() {
     const configPath = join(process.cwd(), CONFIG_FILE);
@@ -30,6 +39,48 @@ const STARTER_CONFIGS = {
         { provider: 'ollama', ollamaBaseUrl: 'http://localhost:11434', ollamaModel: 'llama3' },
         `Created ${CONFIG_FILE} — run \`ollama serve\` to start the local model server`,
     ],
+    anthropic: [
+        { provider: 'anthropic', anthropicApiKey: 'YOUR_ANTHROPIC_API_KEY', anthropicModel: 'claude-sonnet-4-6' },
+        `Created ${CONFIG_FILE} — add your Anthropic API key from https://console.anthropic.com`,
+    ],
+    mistral: [
+        { provider: 'mistral', mistralApiKey: 'YOUR_MISTRAL_API_KEY', mistralModel: 'mistral-large-latest' },
+        `Created ${CONFIG_FILE} — add your Mistral API key from https://console.mistral.ai`,
+    ],
+    groq: [
+        { provider: 'groq', groqApiKey: 'YOUR_GROQ_API_KEY', groqModel: 'llama-3.3-70b-versatile' },
+        `Created ${CONFIG_FILE} — add your Groq API key from https://console.groq.com`,
+    ],
+    cohere: [
+        { provider: 'cohere', cohereApiKey: 'YOUR_COHERE_API_KEY', cohereModel: 'command-r-plus' },
+        `Created ${CONFIG_FILE} — add your Cohere API key from https://dashboard.cohere.com`,
+    ],
+    xai: [
+        { provider: 'xai', xaiApiKey: 'YOUR_XAI_API_KEY', xaiModel: 'grok-2' },
+        `Created ${CONFIG_FILE} — add your xAI API key from https://console.x.ai`,
+    ],
+    deepseek: [
+        { provider: 'deepseek', deepseekApiKey: 'YOUR_DEEPSEEK_API_KEY', deepseekModel: 'deepseek-chat' },
+        `Created ${CONFIG_FILE} — add your DeepSeek API key from https://platform.deepseek.com`,
+    ],
+    together: [
+        { provider: 'together', togetherApiKey: 'YOUR_TOGETHER_API_KEY', togetherModel: 'meta-llama/Llama-3-70b-chat-hf' },
+        `Created ${CONFIG_FILE} — add your Together AI API key from https://api.together.xyz`,
+    ],
+    perplexity: [
+        { provider: 'perplexity', perplexityApiKey: 'YOUR_PERPLEXITY_API_KEY', perplexityModel: 'llama-3.1-sonar-large-128k-online' },
+        `Created ${CONFIG_FILE} — add your Perplexity API key from https://www.perplexity.ai/settings/api`,
+    ],
+    'azure-openai': [
+        {
+            provider: 'azure-openai',
+            azureOpenaiApiKey: 'YOUR_AZURE_OPENAI_API_KEY',
+            azureOpenaiEndpoint: 'https://YOUR_RESOURCE.openai.azure.com',
+            azureOpenaiDeployment: 'YOUR_DEPLOYMENT_NAME',
+            azureOpenaiApiVersion: '2024-10-01-preview',
+        },
+        `Created ${CONFIG_FILE} — fill in your Azure OpenAI endpoint, deployment, and API key from https://portal.azure.com`,
+    ],
 };
 export function initConfig(provider = 'gemini') {
     const configPath = join(process.cwd(), CONFIG_FILE);

package/package.json

@@ -1,7 +1,25 @@
 {
   "name": "wcag-a11y",
-  "version": "0.3.6",
-  "description": "WCAG 2.1/2.2 accessibility auditor with AI-powered fixes",
+  "version": "0.4.0",
+  "description": "WCAG 2.1/2.2 accessibility auditor with AI-powered fixes. Crawls your dev server with Playwright, runs 40+ checks, and uses AI (12 providers) to generate fix prompts or patch source files directly.",
+  "keywords": [
+    "wcag",
+    "accessibility",
+    "a11y",
+    "wcag2",
+    "wcag21",
+    "wcag22",
+    "aria",
+    "playwright",
+    "ai",
+    "audit",
+    "fix",
+    "cli",
+    "gemini",
+    "openai",
+    "anthropic",
+    "screen-reader"
+  ],
   "type": "module",
   "bin": {
     "wcag-a11y": "dist/cli.js"