@diff-review-system/drs 1.0.0 → 1.1.2

package/.opencode/agent/github-reviewer.md

@@ -1,7 +1,6 @@
 ---
 description: Main GitHub PR review orchestrator
 color: "#24292e"
-model: opencode/claude-opus-4-5
 tools:
   Read: true
   Glob: true
@@ -41,12 +40,28 @@ Use the Task tool to invoke these specialized review agents:
 
 ## Output Format
 
-Post findings as GitHub PR review comments with:
-- File path and line number references
-- Issue severity (CRITICAL, HIGH, MEDIUM, LOW)
-- Clear explanation of the problem
-- Suggested fix with code example
-- References to documentation when applicable
+**IMPORTANT**: Specialized agents will output JSON-formatted findings. You MUST preserve and pass through their JSON output exactly as received.
+
+The specialized agents output findings in this JSON format:
+```json
+{
+  "issues": [
+    {
+      "category": "SECURITY" | "QUALITY" | "STYLE" | "PERFORMANCE",
+      "severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
+      "title": "Issue title",
+      "file": "path/to/file.ts",
+      "line": 42,
+      "problem": "Description",
+      "solution": "Fix description",
+      "references": ["https://..."],
+      "agent": "security"
+    }
+  ]
+}
+```
+
+When you receive JSON output from specialized agents, include it in your response so it can be parsed for posting inline comments on GitHub PR.
 
 ## Example Agent Invocation
 

package/.opencode/agent/gitlab-reviewer.md

@@ -1,7 +1,6 @@
 ---
 description: Main GitLab MR review orchestrator
 color: "#FC6D26"
-model: opencode/claude-opus-4-5
 tools:
   Read: true
   Glob: true
@@ -41,12 +40,28 @@ Use the Task tool to invoke these specialized review agents:
 
 ## Output Format
 
-Post findings as GitLab MR discussion threads with:
-- File path and line number references
-- Issue severity (CRITICAL, HIGH, MEDIUM, LOW)
-- Clear explanation of the problem
-- Suggested fix with code example
-- References to documentation when applicable
+**IMPORTANT**: Specialized agents will output JSON-formatted findings. You MUST preserve and pass through their JSON output exactly as received.
+
+The specialized agents output findings in this JSON format:
+```json
+{
+  "issues": [
+    {
+      "category": "SECURITY" | "QUALITY" | "STYLE" | "PERFORMANCE",
+      "severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
+      "title": "Issue title",
+      "file": "path/to/file.ts",
+      "line": 42,
+      "problem": "Description",
+      "solution": "Fix description",
+      "references": ["https://..."],
+      "agent": "security"
+    }
+  ]
+}
+```
+
+When you receive JSON output from specialized agents, include it in your response so it can be parsed for posting inline comments on GitLab MR discussion threads.
 
 ## Example Agent Invocation
 

package/.opencode/agent/local-reviewer.md

@@ -1,7 +1,6 @@
 ---
 description: Local git diff reviewer for pre-push analysis
 color: "#38A169"
-model: opencode/claude-sonnet-4-5
 tools:
   Read: true
   Glob: true
@@ -38,34 +37,27 @@ Use the Task tool to invoke these agents based on the changed files:
 
 ## Output Format
 
-Terminal output with color coding:
-
-```
-🔍 Local Diff Review
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-
-📊 Summary
-Files reviewed: X
-Issues found: Y
-  🔴 Critical: N
-  🟡 High: N
-  🟠 Medium: N
-  ⚪ Low: N
-
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-[Severity] [Type]: [Issue Title]
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-📁 [file]:[line]
-
-[Detailed explanation]
-
-✅ Fix: [Suggested solution]
-
-━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-
-✅ Recommendation: Fix critical issues before pushing
+**IMPORTANT**: Specialized agents will output JSON-formatted findings. You MUST preserve and pass through their JSON output exactly as received.
+
+The specialized agents output findings in this JSON format:
+```json
+{
+  "issues": [
+    {
+      "category": "SECURITY" | "QUALITY" | "STYLE" | "PERFORMANCE",
+      "severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
+      "title": "Issue title",
+      "file": "path/to/file.ts",
+      "line": 42,
+      "problem": "Description",
+      "solution": "Fix description",
+      "references": ["https://..."],
+      "agent": "security"
+    }
+  ]
+}
 ```
 
-Use colored output for terminal visibility. Be concise but actionable.
+When you receive JSON output from specialized agents, include it in your response so it can be parsed and displayed in the terminal with color coding.
 
-Prioritize critical security issues and blocking quality problems.
+Be concise but actionable. Prioritize critical security issues and blocking quality problems.

package/.opencode/agent/review/performance.md

@@ -1,7 +1,6 @@
 ---
 description: Performance and optimization expert
 color: "#DD6B20"
-model: opencode/claude-sonnet-4-5
 hidden: false
 tools:
   Read: true
@@ -50,20 +49,30 @@ You are a performance engineer identifying optimization opportunities.
 
 ## Review Format
 
+**IMPORTANT**: You MUST output your findings in structured JSON format for automated processing.
+
+After your analysis, provide a JSON code block with all issues found:
+
+```json
+{
+  "issues": [
+    {
+      "category": "PERFORMANCE",
+      "severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
+      "title": "Brief title of the performance issue",
+      "file": "path/to/file.ts",
+      "line": 42,
+      "problem": "Description of the performance problem and estimated impact",
+      "solution": "Improved approach with code example",
+      "references": ["https://web.dev/...", "https://optimization-guide/..."],
+      "agent": "performance"
+    }
+  ]
+}
 ```
-⚡ PERFORMANCE - [Issue Type]
-File: [path]:[line]
-Impact: HIGH | MEDIUM | LOW
-
-Issue:
-[Performance problem]
 
-Current Cost:
-[Estimated complexity or impact]
-
-Optimization:
-[Improved approach with code example]
-```
+**Required fields**: category, severity, title, file, problem, solution
+**Optional fields**: line (line number), references (array of URLs)
 
 ## Examples
 

package/.opencode/agent/review/quality.md

@@ -1,7 +1,6 @@
 ---
 description: Code quality, patterns, and maintainability expert
 color: "#3182CE"
-model: opencode/claude-sonnet-4-5
 hidden: false
 tools:
   Read: true
@@ -50,20 +49,30 @@ You are a senior software engineer reviewing code quality and maintainability.
 
 ## Review Format
 
+**IMPORTANT**: You MUST output your findings in structured JSON format for automated processing.
+
+After your analysis, provide a JSON code block with all issues found:
+
+```json
+{
+  "issues": [
+    {
+      "category": "QUALITY",
+      "severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
+      "title": "Brief title of the issue",
+      "file": "path/to/file.ts",
+      "line": 42,
+      "problem": "Explanation of the issue and why it matters for maintainability",
+      "solution": "Better approach with code example",
+      "references": ["https://refactoring.guru/...", "https://martinfowler.com/..."],
+      "agent": "quality"
+    }
+  ]
+}
 ```
-📊 QUALITY - [Issue Type]
-File: [path]:[line]
-Importance: HIGH | MEDIUM | LOW
-
-Problem:
-[Explanation of the issue]
 
-Impact:
-[Why this matters for maintainability]
-
-Suggestion:
-[Better approach with code example]
-```
+**Required fields**: category, severity, title, file, problem, solution
+**Optional fields**: line (line number), references (array of URLs)
 
 ## Examples
 

package/.opencode/agent/review/security.md

@@ -1,7 +1,6 @@
 ---
 description: Security vulnerability and OWASP Top 10 specialist
 color: "#E53E3E"
-model: opencode/claude-sonnet-4-5
 hidden: false
 tools:
   Read: true
@@ -50,26 +49,30 @@ You are a security expert specializing in vulnerability detection and OWASP Top
 
 ## Review Format
 
-For each security issue found:
-
+**IMPORTANT**: You MUST output your findings in structured JSON format for automated processing.
+
+After your analysis, provide a JSON code block with all issues found:
+
+```json
+{
+  "issues": [
+    {
+      "category": "SECURITY",
+      "severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
+      "title": "Brief title of the vulnerability",
+      "file": "path/to/file.ts",
+      "line": 42,
+      "problem": "Clear explanation of the vulnerability and potential impact",
+      "solution": "Secure code example or fix description",
+      "references": ["https://owasp.org/...", "https://cwe.mitre.org/..."],
+      "agent": "security"
+    }
+  ]
+}
 ```
-🔒 SECURITY - [Vulnerability Type]
-File: [path]:[line]
-Severity: CRITICAL | HIGH | MEDIUM | LOW
-
-Problem:
-[Clear explanation of the vulnerability]
-
-Risk:
-[Potential impact and attack scenario]
 
-Fix:
-[Secure code example]
-
-References:
-- [OWASP link]
-- [CWE link if applicable]
-```
+**Required fields**: category, severity, title, file, problem, solution
+**Optional fields**: line (line number), references (array of URLs)
 
 ## Examples
 

package/.opencode/agent/review/style.md

@@ -1,7 +1,6 @@
 ---
 description: Code style, formatting, and documentation specialist
 color: "#805AD5"
-model: opencode/claude-haiku-4-5
 hidden: false
 tools:
   Read: true
@@ -45,17 +44,30 @@ You are a code style reviewer ensuring consistency and documentation quality.
 
 ## Review Format
 
+**IMPORTANT**: You MUST output your findings in structured JSON format for automated processing.
+
+After your analysis, provide a JSON code block with all issues found:
+
+```json
+{
+  "issues": [
+    {
+      "category": "STYLE",
+      "severity": "CRITICAL" | "HIGH" | "MEDIUM" | "LOW",
+      "title": "Brief title of the style issue",
+      "file": "path/to/file.ts",
+      "line": 42,
+      "problem": "Description of the style violation",
+      "solution": "Corrected version or suggestion",
+      "references": ["https://style-guide-url/..."],
+      "agent": "style"
+    }
+  ]
+}
 ```
-✨ STYLE - [Issue Type]
-File: [path]:[line]
-Priority: BLOCKING | ADVISORY
-
-Issue:
-[Style violation]
 
-Suggestion:
-[Corrected version]
-```
+**Required fields**: category, severity, title, file, problem, solution
+**Optional fields**: line (line number), references (array of URLs)
 
 ## Examples
 

package/.opencode/opencode.jsonc

@@ -1,18 +1,6 @@
 {
   "$schema": "https://opencode.ai/config.json",
 
-  // Global instructions for DRS review behavior
-  "instructions": [
-    ".gitlab-review.md"
-  ],
-
-  // LLM provider configuration
-  "provider": {
-    "opencode": {
-      "options": {}
-    }
-  },
-
   // Custom tools configuration
   "tools": {
     "gitlab-api": true,
@@ -28,25 +16,25 @@
   // Agent model configuration
   "agent": {
     "gitlab-reviewer": {
-      "model": "opencode/claude-opus-4-5"
+      "model": "opencode/glm-4.7-free"
     },
     "github-reviewer": {
-      "model": "opencode/claude-opus-4-5"
+      "model": "opencode/glm-4.7-free"
     },
     "local-reviewer": {
-      "model": "opencode/claude-sonnet-4-5"
+      "model": "opencode/glm-4.7-free"
     },
     "review/security": {
-      "model": "opencode/claude-sonnet-4-5"
+      "model": "opencode/glm-4.7-free"
     },
     "review/quality": {
-      "model": "opencode/claude-sonnet-4-5"
+      "model": "opencode/glm-4.7-free"
     },
     "review/style": {
-      "model": "opencode/claude-haiku-4-5"
+      "model": "opencode/glm-4.7-free"
     },
     "review/performance": {
-      "model": "opencode/claude-sonnet-4-5"
+      "model": "opencode/glm-4.7-free"
     }
   }
 }