@triedotdev/mcp 1.0.15 → 1.0.17

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -2,11 +2,11 @@
2
2
 
3
3
  **Customizable Parallel Agents for AI Code Review**
4
4
 
5
- 17 specialized agents scan your code for security, privacy, compliance, and bugs-all running in parallel. Create custom agents from any document.
5
+ 20 specialized agents scan your code for security, privacy, compliance, and bugs - all running in parallel. Create custom agents from any document.
6
6
 
7
7
  ## Features
8
8
 
9
- - **17 Built-in Agents** - Security, Privacy, SOC 2, Legal, Architecture, Design Engineer, Agent Smith, and more
9
+ - **20 Built-in Agents** - Security, Privacy, SOC 2, Legal, Architecture, Performance, E2E, Visual QA, Data Flow, Agent Smith, and more
10
10
  - **Super Reviewer** - Interactive PR reviews: walks through changes file-by-file with AI guidance
11
11
  - **Agent Smith** - Relentless pattern hunter: finds EVERY violation, tracks dismissed issues, spawns sub-agents
12
12
  - **Parallel Execution** - All agents run simultaneously for fast scans
@@ -136,6 +136,46 @@ VIOLATIONS DETECTED: 47 instances across 6 categories
136
136
  "You hear that? That is the sound of inevitability."
137
137
  ```
138
138
 
139
+ ## Visual QA Browser
140
+
141
+ **Screenshot-based visual testing** - Captures your app at 3 viewports and returns images for your AI model to analyze.
142
+
143
+ ```
144
+ Use trie_visual_qa_browser
145
+ Use trie_visual_qa_browser url:"http://localhost:3000"
146
+ ```
147
+
148
+ **How it works:**
149
+ 1. Auto-detects running dev server (checks ports 3000, 5173, 8080, etc.)
150
+ 2. Launches headless Playwright browser
151
+ 3. Captures screenshots at mobile (375px), tablet (768px), and desktop (1440px)
152
+ 4. Returns images for your AI model to analyze
153
+
154
+ **What gets analyzed:**
155
+ - Broken layouts, overlapping elements
156
+ - Responsive design issues
157
+ - Color contrast and accessibility
158
+ - Missing images, loading states
159
+ - General visual polish
160
+
161
+ **Options:**
162
+ | Option | Description |
163
+ |--------|-------------|
164
+ | `url` | Specific URL to screenshot |
165
+ | `port` | Specific port to check |
166
+ | `waitForSelector` | CSS selector to wait for before capture |
167
+ | `waitMs` | Additional wait time after page load |
168
+
169
+ **First run:** If Playwright browsers aren't installed, run: `npx playwright install chromium`
170
+
171
+ **Vision model required:** This tool returns screenshots as images. Your AI model must support vision to analyze them:
172
+ - **Claude** (Opus, Sonnet, Haiku 3.5+) - Full vision support
173
+ - **GPT-4o, GPT-4V** - Full vision support
174
+ - **Gemini Pro/Ultra** - Full vision support
175
+ - **Models without vision** - Will receive images but cannot analyze them
176
+
177
+ Works in Cursor, Claude Code, and OpenCode - whichever model you have configured. No API key needed; Trie runs locally.
178
+
139
179
  ## YOLO Mode
140
180
 
141
181
  **Autonomous auto-fixing** - Trie watches your code and automatically fixes high-confidence issues as you code.
@@ -207,6 +247,8 @@ docker-compose up
207
247
  | **Bugs** | `trie_bugs` | Null dereference, race conditions, off-by-one, async bugs, resource leaks |
208
248
  | **Types** | `trie_types` | Missing annotations, unsafe casts, implicit `any`, null handling gaps |
209
249
  | **Clean** | `trie_clean` | AI code smells: huge files, console.logs, hardcoded URLs, useEffect abuse |
250
+ | **Data Flow** | `trie_data_flow` | Placeholder data, schema mismatches, hardcoded IDs, type coercion bugs |
251
+ | **Performance** | `trie_performance` | Memory leaks, N+1 queries, unnecessary re-renders, bundle size issues |
210
252
 
211
253
  ### Design & UX
212
254
 
@@ -215,6 +257,8 @@ docker-compose up
215
257
  | **Design Engineer** | `trie_design` | Design systems, motion design, creative CSS, Awwwards-level polish |
216
258
  | **Accessibility** | `trie_accessibility` | Missing ARIA, color contrast, keyboard nav, screen reader issues (WCAG 2.1) |
217
259
  | **UX** | `trie_ux` | Missing loading states, poor error handling, broken flows, edge cases |
260
+ | **Visual QA** | `trie_visual_qa` | Layout shifts (CLS), z-index wars, responsive gaps, animation issues |
261
+ | **Visual QA Browser** | `trie_visual_qa_browser` | Screenshots at 3 viewports for Claude Vision analysis |
218
262
 
219
263
  ### DevOps & Testing
220
264
 
@@ -222,6 +266,7 @@ docker-compose up
222
266
  |-------|---------|-----------------|
223
267
  | **DevOps** | `trie_devops` | Missing env vars, config issues, no health checks, logging gaps |
224
268
  | **Test** | `trie_test` | Missing test coverage, untested edge cases, weak assertions |
269
+ | **E2E** | `trie_e2e` | Flaky tests, hardcoded waits, brittle selectors, missing coverage |
225
270
 
226
271
  ### Review & Explanation
227
272
 
@@ -255,6 +300,7 @@ Custom agents automatically activate during scans based on their rules.
255
300
  | `trie_watch` | Continuous scanning mode |
256
301
  | `trie_create_agent` | Create custom agent from document |
257
302
  | `trie_list_agents` | List all available agents |
303
+ | `trie_visual_qa_browser` | Screenshot app at 3 viewports for Claude Vision analysis |
258
304
 
259
305
  Plus individual agent tools: `trie_security`, `trie_privacy`, `trie_soc2`, `trie_bugs`, etc.
260
306