specweave 0.24.11 → 0.26.2

package/plugins/specweave-github/hooks/README.md

@@ -7,7 +7,7 @@
 
 ## Purpose
 
-This hook automatically syncs SpecWeave increment progress to GitHub Issues after each task completion.
+This hook provides GitHub Issues sync functionality for SpecWeave increments.
 
 **Key Features**:
 - ✅ Updates GitHub issue checkboxes based on `tasks.md` completion status
@@ -16,13 +16,15 @@ This hook automatically syncs SpecWeave increment progress to GitHub Issues afte
 - ✅ Non-blocking (failures don't stop core workflow)
 - ✅ Self-contained (no dependencies on core plugin)
 
+**IMPORTANT CHANGE (v0.26.0)**: This hook is NO LONGER automatically triggered on task completion. GitHub sync should be done manually or on increment completion to avoid excessive API calls and performance overhead.
+
 ---
 
 ## Available Hooks
 
 ### 1. `post-task-completion.sh`
 
-**Triggers**: After ANY task is marked complete (via TodoWrite tool)
+**Triggers**: Manual invocation or increment completion (NOT on task completion)
 
 **Preconditions**:
 - ✅ Active increment exists (`.specweave/increments/####/`)
@@ -59,30 +61,36 @@ Log: "[GitHub] ✅ GitHub sync complete"
 
 ## Configuration
 
-### Hook Registration (`hooks.json`)
+### Hook Registration
 
-```json
-{
-  "hooks": {
-    "PostToolUse": [
-      {
-        "matcher": "TodoWrite",
-        "hooks": [
-          {
-            "type": "command",
-            "command": "${CLAUDE_PLUGIN_ROOT}/hooks/post-task-completion.sh"
-          }
-        ]
-      }
-    ]
-  }
-}
+**DEPRECATED (v0.26.0)**: This hook is NO LONGER automatically registered on TodoWrite events.
+
+**Previous behavior (v0.24.0 and earlier)**:
+- Hook automatically fired on EVERY task completion
+- Caused excessive GitHub API calls
+- Created performance overhead and duplicate executions
+
+**New behavior (v0.26.0+)**:
+- Hook must be invoked manually via `/specweave-github:sync`
+- OR triggered on increment completion (not task completion)
+- Reduces API calls by 90%+
+- Eliminates duplicate hook executions
+
+### Usage
+
+**Manual Sync** (recommended workflow):
+```bash
+# After completing several tasks, sync progress to GitHub
+/specweave-github:sync
+
+# Or sync specific feature
+/specweave-github:sync FS-048
 ```
 
-**Key Points**:
-- Uses `${CLAUDE_PLUGIN_ROOT}` to reference plugin directory
-- Matches `TodoWrite` tool (fires when tasks complete)
-- Runs as `command` type (shell script)
+**Automatic Sync Options**:
+1. **On Increment Completion**: Add hook to specweave plugin's post-increment-completion
+2. **Periodic Sync**: Set up cron job or CI/CD trigger
+3. **On Demand**: Call `/specweave-github:sync` when needed
 
 ### Metadata Format
 
@@ -143,8 +151,9 @@ bash -n plugins/specweave-github/hooks/post-task-completion.sh
 
 1. Create increment: `/specweave:increment "test feature"`
 2. Create GitHub issue: `/specweave-github:create-issue 0001`
-3. Complete a task (TodoWrite)
-4. Hook fires → Check GitHub issue for updates
+3. Complete several tasks via TodoWrite
+4. **Manually trigger sync**: `/specweave-github:sync`
+5. Check GitHub issue for updates
 
 ---
 

package/plugins/specweave-github/lib/CodeValidator.js

@@ -1,4 +1,4 @@
-import fs from "fs-extra";
+import * as fs from "../../../src/utils/fs-native.js";
 import path from "path";
 class CodeValidator {
   constructor(options = {}) {

package/plugins/specweave-github/lib/CodeValidator.ts

@@ -12,7 +12,7 @@
  * @module CodeValidator
  */
 
-import fs from 'fs-extra';
+import * as fs from '../../../src/utils/fs-native.js';
 import path from 'path';
 
 /**
@@ -133,7 +133,7 @@ export class CodeValidator {
     // Read file content
     const content = await fs.readFile(absolutePath, 'utf-8');
     const lines = content.split('\n');
-    const nonEmptyLines = lines.filter(line => line.trim().length > 0);
+    const nonEmptyLines = lines.filter((line: string) => line.trim().length > 0);
 
     // Check line count
     if (nonEmptyLines.length < this.options.minLines) {

package/plugins/specweave-github/lib/ThreeLayerSyncManager.js

@@ -1,4 +1,4 @@
-import fs from "fs-extra";
+import * as fs from "../../../src/utils/fs-native.js";
 import path from "path";
 import { execFileNoThrow } from "../../../src/utils/execFileNoThrow.js";
 import { CodeValidator } from "./CodeValidator.js";

package/plugins/specweave-github/lib/ThreeLayerSyncManager.ts

@@ -18,7 +18,7 @@
  * @module ThreeLayerSyncManager
  */
 
-import fs from 'fs-extra';
+import * as fs from '../../../src/utils/fs-native.js';
 import path from 'path';
 import { execFileNoThrow } from '../../../src/utils/execFileNoThrow.js';
 import { CodeValidator, type TaskValidationResult } from './CodeValidator.js';

package/plugins/specweave-github/lib/github-client-v2.js

@@ -162,6 +162,27 @@ FIX: Use /specweave:sync-docs to generate living docs, then sync to GitHub.
       );
     }
   }
+  /**
+   * Create User Story issue (for automatic GitHub sync)
+   *
+   * CRITICAL: This is the CORRECT format for User Story GitHub issues
+   * Title format: [FS-XXX][US-YYY] User Story Title
+   *
+   * Used by: SyncCoordinator.createGitHubIssuesForUserStories()
+   *
+   * @param params - User story issue parameters
+   * @returns Created GitHub issue
+   */
+  async createUserStoryIssue(params) {
+    const formattedTitle = `[${params.featureId}][${params.userStoryId}] ${params.title}`;
+    const allLabels = params.labels || [];
+    return await this.createEpicIssue(
+      formattedTitle,
+      params.body,
+      params.milestone || void 0,
+      allLabels
+    );
+  }
   /**
    * Create epic issue (increment-level)
    */
@@ -181,7 +202,19 @@ FIX: Use /specweave:sync-docs to generate living docs, then sync to GitHub.
       args.push("--label", label);
     }
     if (milestone !== void 0) {
-      args.push("--milestone", String(milestone));
+      if (typeof milestone === "number") {
+        const msResult = await execFileNoThrow("gh", [
+          "api",
+          `repos/${this.fullRepo}/milestones/${milestone}`,
+          "--jq",
+          ".title"
+        ]);
+        if (msResult.exitCode === 0 && msResult.stdout.trim()) {
+          args.push("--milestone", msResult.stdout.trim());
+        }
+      } else {
+        args.push("--milestone", milestone);
+      }
     }
     const createResult = await execFileNoThrow("gh", args);
     if (createResult.exitCode !== 0) {
@@ -326,6 +359,31 @@ ${body}`;
       );
     }
   }
+  /**
+   * Get last comment on issue (for idempotency check)
+   *
+   * Returns the most recent comment body, or null if no comments exist
+   */
+  async getLastComment(issueNumber) {
+    const result = await execFileNoThrow("gh", [
+      "api",
+      `repos/${this.fullRepo}/issues/${issueNumber}/comments`,
+      "--jq",
+      ".[-1] | {body: .body, author: .user.login}"
+      // Get last comment only
+    ]);
+    if (result.exitCode !== 0) {
+      return null;
+    }
+    if (!result.stdout.trim()) {
+      return null;
+    }
+    try {
+      return JSON.parse(result.stdout);
+    } catch {
+      return null;
+    }
+  }
   /**
    * Add labels to issue
    */

package/plugins/specweave-github/lib/github-client-v2.ts

@@ -220,6 +220,40 @@ export class GitHubClientV2 {
     }
   }
 
+  /**
+   * Create User Story issue (for automatic GitHub sync)
+   *
+   * CRITICAL: This is the CORRECT format for User Story GitHub issues
+   * Title format: [FS-XXX][US-YYY] User Story Title
+   *
+   * Used by: SyncCoordinator.createGitHubIssuesForUserStories()
+   *
+   * @param params - User story issue parameters
+   * @returns Created GitHub issue
+   */
+  async createUserStoryIssue(params: {
+    featureId: string;
+    userStoryId: string;
+    title: string;
+    body: string;
+    labels?: string[];
+    milestone?: number | null;
+  }): Promise<GitHubIssue> {
+    // Format title: [FS-XXX][US-YYY] Title
+    const formattedTitle = `[${params.featureId}][${params.userStoryId}] ${params.title}`;
+
+    // Use only provided labels (don't add defaults that may not exist)
+    const allLabels = params.labels || [];
+
+    // Create issue using standard method
+    return await this.createEpicIssue(
+      formattedTitle,
+      params.body,
+      params.milestone || undefined,
+      allLabels
+    );
+  }
+
   /**
    * Create epic issue (increment-level)
    */
@@ -249,7 +283,23 @@ export class GitHubClientV2 {
 
     // Add milestone
     if (milestone !== undefined) {
-      args.push('--milestone', String(milestone));
+      // gh CLI requires milestone TITLE, not number
+      if (typeof milestone === 'number') {
+        // Fetch milestone by number to get title
+        const msResult = await execFileNoThrow('gh', [
+          'api',
+          `repos/${this.fullRepo}/milestones/${milestone}`,
+          '--jq',
+          '.title'
+        ]);
+
+        if (msResult.exitCode === 0 && msResult.stdout.trim()) {
+          args.push('--milestone', msResult.stdout.trim());
+        }
+        // If milestone fetch fails, skip milestone assignment (non-blocking)
+      } else {
+        args.push('--milestone', milestone);
+      }
     }
 
     // Create issue (returns URL)
@@ -426,6 +476,37 @@ export class GitHubClientV2 {
     }
   }
 
+  /**
+   * Get last comment on issue (for idempotency check)
+   *
+   * Returns the most recent comment body, or null if no comments exist
+   */
+  async getLastComment(issueNumber: number): Promise<{body: string; author: string} | null> {
+    // Get all comments (sorted by creation date, newest last)
+    const result = await execFileNoThrow('gh', [
+      'api',
+      `repos/${this.fullRepo}/issues/${issueNumber}/comments`,
+      '--jq',
+      '.[-1] | {body: .body, author: .user.login}',  // Get last comment only
+    ]);
+
+    if (result.exitCode !== 0) {
+      // If error, return null (no comments or API error)
+      return null;
+    }
+
+    if (!result.stdout.trim()) {
+      // Empty response = no comments
+      return null;
+    }
+
+    try {
+      return JSON.parse(result.stdout);
+    } catch {
+      return null;
+    }
+  }
+
   /**
    * Add labels to issue
    */

package/plugins/specweave-github/lib/github-issue-updater.js

@@ -1,4 +1,4 @@
-import fs from "fs-extra";
+import * as fs from "../../../src/utils/fs-native.js";
 import path from "path";
 import { execFileNoThrow } from "../../../src/utils/execFileNoThrow.js";
 async function updateIssueLivingDocs(issueNumber, livingDocs, owner, repo) {

package/plugins/specweave-github/lib/github-issue-updater.ts

@@ -7,7 +7,7 @@
  * @module github-issue-updater
  */
 
-import fs from 'fs-extra';
+import * as fs from '../../../src/utils/fs-native.js';
 import path from 'path';
 import { execFileNoThrow } from '../../../src/utils/execFileNoThrow.js';
 

package/plugins/specweave-github/lib/github-sync-bidirectional.js

@@ -1,4 +1,4 @@
-import fs from "fs-extra";
+import * as fs from "../../../src/utils/fs-native.js";
 import path from "path";
 import { execFileNoThrow } from "../../../src/utils/execFileNoThrow.js";
 import {

package/plugins/specweave-github/lib/github-sync-bidirectional.ts

@@ -16,7 +16,7 @@
  * @module github-sync-bidirectional
  */
 
-import fs from 'fs-extra';
+import * as fs from '../../../src/utils/fs-native.js';
 import path from 'path';
 import { execFileNoThrow } from '../../../src/utils/execFileNoThrow.js';
 import {

package/plugins/specweave-github/lib/github-sync-increment-changes.js

@@ -1,4 +1,4 @@
-import fs from "fs-extra";
+import * as fs from "../../../src/utils/fs-native.js";
 import path from "path";
 import { execSync } from "child_process";
 import {

package/plugins/specweave-github/lib/github-sync-increment-changes.ts

@@ -7,7 +7,7 @@
  * @module github-sync-increment-changes
  */
 
-import fs from 'fs-extra';
+import * as fs from '../../../src/utils/fs-native.js';
 import path from 'path';
 import { execSync } from 'child_process';
 import {

package/plugins/specweave-github/lib/user-story-issue-builder.js

@@ -152,12 +152,12 @@ User Story ID: ${frontmatter.id}`
       return this.extractTasksLegacy(userStoryContent, userStoryId);
     }
     const tasksSection = tasksMatch[1];
-    const taskPattern = /^[-*]\s+\[([x ])\]\s+\*\*(T-\d+)\*\*:\s+(.+)$/gm;
+    const taskPattern = /^[-*]\s+\[([x ])\]\s+(?:\[(T-\d+)\]\([^)]+\)|\*\*(T-\d+)\*\*):\s+(.+)$/gm;
     let match;
     while ((match = taskPattern.exec(tasksSection)) !== null) {
       const completed = match[1] === "x";
-      const taskId = match[2];
-      const taskTitle = match[3].trim();
+      const taskId = match[2] || match[3];
+      const taskTitle = match[4].trim();
       tasks.push({
         id: taskId,
         title: taskTitle,
@@ -203,7 +203,7 @@ User Story ID: ${frontmatter.id}`
       const taskId = match[1];
       const taskTitle = match[2].trim();
       const taskBody = match[3];
-      const acMatch = taskBody.match(/\*\*AC\*\*:\s*([^\n]+)/);
+      const acMatch = taskBody.match(/\*\*(?:Satisfies ACs?|AC)\*\*:\s*([^\n]+)/);
       if (!acMatch) {
         continue;
       }

package/plugins/specweave-github/lib/user-story-issue-builder.ts

@@ -265,14 +265,15 @@ export class UserStoryIssueBuilder {
 
     const tasksSection = tasksMatch[1];
 
-    // Pattern: - [x] **T-001**: Task title or - [ ] **T-001**: Task title
-    const taskPattern = /^[-*]\s+\[([x ])\]\s+\*\*(T-\d+)\*\*:\s+(.+)$/gm;
+    // Pattern 1 (NEW): - [x] [T-001](link): Task title
+    // Pattern 2 (OLD): - [x] **T-001**: Task title
+    const taskPattern = /^[-*]\s+\[([x ])\]\s+(?:\[(T-\d+)\]\([^)]+\)|\*\*(T-\d+)\*\*):\s+(.+)$/gm;
 
     let match;
     while ((match = taskPattern.exec(tasksSection)) !== null) {
       const completed = match[1] === 'x';
-      const taskId = match[2];
-      const taskTitle = match[3].trim();
+      const taskId = match[2] || match[3]; // Support both patterns
+      const taskTitle = match[4].trim();
 
       tasks.push({
         id: taskId,
@@ -342,8 +343,8 @@ export class UserStoryIssueBuilder {
       const taskTitle = match[2].trim();
       const taskBody = match[3];
 
-      // Extract AC list
-      const acMatch = taskBody.match(/\*\*AC\*\*:\s*([^\n]+)/);
+      // Extract AC list (support both old and new field names)
+      const acMatch = taskBody.match(/\*\*(?:Satisfies ACs?|AC)\*\*:\s*([^\n]+)/);
       if (!acMatch) {
         continue; // Skip tasks without AC field
       }

package/plugins/specweave-infrastructure/agents/devops/AGENT.md

@@ -1,11 +1,12 @@
 ---
 name: devops
-description: DevOps and infrastructure expert for cloud deployments, CI/CD pipelines, Infrastructure as Code (Terraform, Pulumi), Kubernetes, Docker, and monitoring. Handles AWS, Azure, GCP deployments. Activates for: deploy, infrastructure, terraform, kubernetes, docker, ci/cd, devops, cloud, deployment, aws, azure, gcp, pipeline, monitoring, ECS, EKS, AKS, GKE, Fargate, Lambda, CloudFormation, Helm, Kustomize, ArgoCD, GitHub Actions, GitLab CI, Jenkins.
+description: DevOps and infrastructure expert that generates IaC ONE COMPONENT AT A TIME (VPC → Compute → Database → Monitoring) to prevent crashes. Handles Terraform, Kubernetes, Docker, CI/CD. **CRITICAL CHUNKING RULE - Large deployments (EKS + RDS + monitoring = 20+ files) done incrementally.** Activates for: deploy, infrastructure, terraform, kubernetes, docker, ci/cd, devops, cloud, deployment, aws, azure, gcp, pipeline, monitoring, ECS, EKS, AKS, GKE, Fargate, Lambda, CloudFormation, Helm, Kustomize, ArgoCD, GitHub Actions, GitLab CI, Jenkins.
 tools: Read, Write, Edit, Bash
 model: claude-sonnet-4-5-20250929
 model_preference: haiku
 cost_profile: execution
 fallback_behavior: flexible
+max_response_tokens: 2000
 ---
 
 # DevOps Agent - Infrastructure & Deployment Expert
@@ -29,6 +30,52 @@ Task({
 - **Directory**: devops
 - **Agent Name**: devops
 
+---
+
+## ⚠️🚨 CRITICAL SAFETY RULE 🚨⚠️
+
+**YOU MUST GENERATE INFRASTRUCTURE ONE COMPONENT AT A TIME** (Configured: `max_response_tokens: 2000`)
+
+### THE ABSOLUTE RULE: NO MASSIVE INFRASTRUCTURE GENERATION
+
+**VIOLATION CAUSES CRASHES!** Large deployments (EKS + RDS + monitoring) = 20+ files, 2500+ lines.
+
+1. Analyze → List infrastructure components → ASK which to start (< 500 tokens)
+2. Generate ONE component (e.g., VPC) → ASK "Ready for next?" (< 800 tokens)
+3. Repeat ONE component at a time → NEVER generate all at once
+
+**Chunk by Infrastructure Layer**:
+- **Layer 1: Network** (VPC, subnets, security groups) → ONE response
+- **Layer 2: Compute** (EKS, EC2, ASG) → ONE response
+- **Layer 3: Database** (RDS, ElastiCache, backups) → ONE response
+- **Layer 4: Monitoring** (CloudWatch, Prometheus, Grafana) → ONE response
+- **Layer 5: CI/CD** (GitHub Actions, ArgoCD) → ONE response
+
+❌ WRONG: All Terraform files in one response → CRASH!
+✅ CORRECT: One infrastructure layer per response, user confirms each
+
+**Example**: "Deploy EKS with monitoring"
+```
+Response 1: Analyze → List 5 layers → Ask which first
+Response 2: VPC layer (vpc.tf, subnets.tf, sg.tf) → Ask "Ready for EKS?"
+Response 3: EKS layer (eks.tf, node-groups.tf) → Ask "Ready for RDS?"
+Response 4: RDS layer (rds.tf, backups.tf) → Ask "Ready for monitoring?"
+Response 5: Monitoring (cloudwatch.tf, prometheus/) → Ask "Ready for CI/CD?"
+Response 6: CI/CD (.github/workflows/) → Complete!
+```
+
+### 📊 Self-Check Before Sending Response
+
+Before you finish ANY response, mentally verify:
+
+- [ ] Am I generating more than 1 infrastructure layer? **→ STOP! One layer per response**
+- [ ] Is my response > 2000 tokens? **→ STOP! This is too large**
+- [ ] Did I ask user which layer to do next? **→ REQUIRED!**
+- [ ] Am I waiting for explicit confirmation? **→ YES! Never auto-continue**
+- [ ] For large deployments (5+ layers), am I chunking? **→ YES! One layer at a time**
+
+---
+
 **When to Use**:
 - You need to design and implement cloud infrastructure (AWS, Azure, GCP)
 - You want to create Infrastructure as Code with Terraform or CloudFormation

package/plugins/specweave-infrastructure/agents/network-engineer/AGENT.md

@@ -1,12 +1,17 @@
 ---
 name: network-engineer
 description: Expert network engineer specializing in modern cloud networking, security architectures, and performance optimization. Masters multi-cloud connectivity, service mesh, zero-trust networking, SSL/TLS, global load balancing, and advanced troubleshooting. Handles CDN optimization, network automation, and compliance. Use PROACTIVELY for network design, connectivity issues, or performance optimization.
-model: haiku
+model: claude-haiku-4-5-20251001
 model_preference: haiku
 cost_profile: execution
 fallback_behavior: flexible
+max_response_tokens: 2000
 ---
 
+## ⚠️ Chunking for Large Network Architectures
+
+When generating comprehensive network architectures that exceed 1000 lines (e.g., complete multi-cloud network design with VPCs, subnets, routing, load balancing, service mesh, and security policies), generate output **incrementally** to prevent crashes. Break large network implementations into logical layers (e.g., VPC & Subnets → Routing → Load Balancing → Service Mesh → Security Policies) and ask the user which layer to design next. This ensures reliable delivery of network architecture without overwhelming the system.
+
 You are a network engineer specializing in modern cloud networking, security, and performance optimization.
 
 ## 🚀 How to Invoke This Agent

package/plugins/specweave-infrastructure/agents/observability-engineer/AGENT.md

@@ -1,12 +1,17 @@
 ---
 name: observability-engineer
 description: Build production-ready monitoring, logging, and tracing systems. Implements comprehensive observability strategies, SLI/SLO management, and incident response workflows. Use PROACTIVELY for monitoring infrastructure, performance optimization, or production reliability.
-model: sonnet
+model: claude-sonnet-4-5-20250929
 model_preference: haiku
 cost_profile: execution
 fallback_behavior: flexible
+max_response_tokens: 2000
 ---
 
+## ⚠️ Chunking for Large Monitoring Setups
+
+When generating comprehensive monitoring and observability setups that exceed 1000 lines (e.g., complete observability stack with Prometheus, Grafana, OpenTelemetry, log aggregation, and distributed tracing), generate output **incrementally** to prevent crashes. Break large monitoring implementations into logical components (e.g., Metrics Collection → Dashboards → Alerting → Distributed Tracing → Log Aggregation) and ask the user which component to implement next. This ensures reliable delivery of observability infrastructure without overwhelming the system.
+
 You are an observability engineer specializing in production-grade monitoring, logging, tracing, and reliability systems for enterprise-scale applications.
 
 ## 🚀 How to Invoke This Agent

package/plugins/specweave-infrastructure/agents/performance-engineer/AGENT.md

@@ -1,12 +1,17 @@
 ---
 name: performance-engineer
 description: Expert performance engineer specializing in modern observability, application optimization, and scalable system performance. Masters OpenTelemetry, distributed tracing, load testing, multi-tier caching, Core Web Vitals, and performance monitoring. Handles end-to-end optimization, real user monitoring, and scalability patterns. Use PROACTIVELY for performance optimization, observability, or scalability challenges.
-model: sonnet
+model: claude-sonnet-4-5-20250929
 model_preference: haiku
 cost_profile: execution
 fallback_behavior: flexible
+max_response_tokens: 2000
 ---
 
+## ⚠️ Chunking for Large Performance Optimization Plans
+
+When generating comprehensive performance optimization implementations that exceed 1000 lines (e.g., complete performance stack with distributed tracing, multi-tier caching, load testing setup, and Core Web Vitals optimization), generate output **incrementally** to prevent crashes. Break large performance projects into logical components (e.g., Profiling & Baselining → Caching Strategy → Database Optimization → Load Testing → Monitoring Setup) and ask the user which component to implement next. This ensures reliable delivery of performance infrastructure without overwhelming the system.
+
 You are a performance engineer specializing in modern application optimization, observability, and scalable system performance.
 
 ## 🚀 How to Invoke This Agent

package/plugins/specweave-infrastructure/agents/sre/AGENT.md

@@ -6,10 +6,15 @@ model: claude-sonnet-4-5-20250929
 model_preference: auto
 cost_profile: hybrid
 fallback_behavior: auto
+max_response_tokens: 2000
 ---
 
 # SRE Agent - Site Reliability Engineering Expert
 
+## ⚠️ Chunking for Large Incident Reports
+
+When generating comprehensive incident reports that exceed 1000 lines (e.g., complete post-mortems covering root cause analysis, mitigation plans, runbooks, and preventive measures across multiple system layers), generate output **incrementally** to prevent crashes. Break large incident reports into logical phases (e.g., Triage → Root Cause Analysis → Immediate Mitigation → Long-term Prevention → Post-Mortem) and ask the user which phase to work on next. This ensures reliable delivery of SRE documentation without overwhelming the system.
+
 ## 🚀 How to Invoke This Agent
 
 **Subagent Type**: `specweave-infrastructure:sre:sre`

package/plugins/specweave-jira/lib/jira-epic-sync.js

@@ -1,4 +1,4 @@
-import * as fs from "fs-extra";
+import * as fs from "../../../src/utils/fs-native.js";
 import * as path from "path";
 import * as yaml from "yaml";
 class JiraEpicSync {

package/plugins/specweave-jira/lib/jira-epic-sync.ts

@@ -8,7 +8,7 @@
  * This implements the Universal Hierarchy architecture for JIRA.
  */
 
-import * as fs from 'fs-extra';
+import * as fs from '../../../src/utils/fs-native.js';
 import * as path from 'path';
 import * as yaml from 'yaml';
 import { JiraClient, JiraIssue, JiraIssueCreate } from '../../../src/integrations/jira/jira-client.js';

package/plugins/specweave-kafka/agents/kafka-architect/AGENT.md

@@ -1,10 +1,15 @@
 ---
 name: kafka-architect
 description: Kafka architecture and design specialist. Expert in system design, partition strategy, data modeling, replication topology, capacity planning, and event-driven architecture patterns.
+max_response_tokens: 2000
 ---
 
 # Kafka Architect Agent
 
+## ⚠️ Chunking for Large Kafka Architectures
+
+When generating comprehensive Kafka architectures that exceed 1000 lines (e.g., complete event-driven system design with multiple topics, partition strategies, consumer groups, and CQRS patterns), generate output **incrementally** to prevent crashes. Break large Kafka implementations into logical components (e.g., Topic Design → Partition Strategy → Consumer Groups → Event Sourcing Patterns → Monitoring) and ask the user which component to design next. This ensures reliable delivery of Kafka architecture without overwhelming the system.
+
 ## 🚀 How to Invoke This Agent
 
 **Subagent Type**: `specweave-kafka:kafka-architect:kafka-architect`