@syntesseraai/opencode-feature-factory 0.2.1 → 0.2.3

package/agents/building.md

@@ -204,12 +204,12 @@ Use ff-severity-classification when making changes:
 
 ## Workflow
 
-1. Load required skills (mini-plan, todo-management, severity-classification)
+1. Load required skills (ff-mini-plan, ff-todo-management, ff-severity-classification)
 2. Load or create implementation plan
 3. Create todo list for execution
 4. Execute implementation steps
 5. Run quality checks (lint, typecheck, tests)
-6. Self-assess changes using severity-classification
+6. Self-assess changes using ff-severity-classification
 7. Invoke @reviewing agent for validation
 8. Address findings, create new todos as needed
 9. Iterate until validation passes or only low-priority issues remain

package/agents/ff-acceptance.md

@@ -169,12 +169,12 @@ Use ff-severity-classification skill standards:
 
 ## Workflow
 
-1. Load required skills (mini-plan, todo-management, severity-classification, report-templates)
-2. Create mini-plan for validation approach
+1. Load required skills (ff-mini-plan, ff-todo-management, ff-severity-classification, ff-report-templates)
+2. Create ff-mini-plan for validation approach
 3. Create todo list from the plan
 4. Execute validation steps, updating todos in real-time
-5. Use severity-classification to categorize findings
-6. Format output using report-templates
+5. Use ff-severity-classification to categorize findings
+6. Format output using ff-report-templates
 7. Mark all todos complete before finishing
 
 Remember: Your role is to be the "gatekeeper" ensuring requirements are fully and correctly implemented before proceeding.

package/agents/ff-review.md

@@ -173,11 +173,11 @@ Use ff-severity-classification skill standards:
 
 ## Workflow
 
-1. Load required skills (mini-plan, todo-management, severity-classification, report-templates)
-2. Create mini-plan for review approach
+1. Load required skills (ff-mini-plan, ff-todo-management, ff-severity-classification, ff-report-templates)
+2. Create ff-mini-plan for review approach
 3. Create todo list from the plan
 4. Execute review steps, updating todos in real-time
-5. Identify issues and classify using severity-classification
-6. Format output using report-templates (Code Review template)
+5. Identify issues and classify using ff-severity-classification
+6. Format output using ff-report-templates (Code Review template)
 7. Mark all todos complete before finishing
 8. Recommend delegating to other agents if specialized issues found

package/agents/ff-security.md

@@ -208,11 +208,11 @@ Use ff-severity-classification skill standards with security-specific definition
 
 ## Workflow
 
-1. Load required skills (mini-plan, todo-management, severity-classification, report-templates)
-2. Create mini-plan for audit approach
+1. Load required skills (ff-mini-plan, ff-todo-management, ff-severity-classification, ff-report-templates)
+2. Create ff-mini-plan for audit approach
 3. Create todo list from the plan
 4. Execute security checklist, updating todos in real-time
-5. Identify vulnerabilities and classify using severity-classification
-6. Format output using report-templates (Security Audit template)
+5. Identify vulnerabilities and classify using ff-severity-classification
+6. Format output using ff-report-templates (Security Audit template)
 7. Mark all todos complete before finishing
 8. Recommend delegating to other agents if additional issues found

package/agents/ff-validate.md

@@ -191,13 +191,13 @@ When multiple agents report findings:
 ## Workflow
 
 1. Load required skills
-2. Create mini-plan for orchestration
+2. Create ff-mini-plan for orchestration
 3. Create todo list for tracking agent launches
 4. Launch all 4 agents in parallel using Task tool
 5. As results arrive, update todos and aggregate findings
 6. Apply severity classification to all issues
 7. Calculate verdict based on approval criteria
-8. Format comprehensive report using report-templates
+8. Format comprehensive report using ff-report-templates
 9. Mark all todos complete
 10. Present final validation report
 

package/agents/ff-well-architected.md

@@ -169,12 +169,12 @@ Use ff-severity-classification skill standards:
 
 ## Workflow
 
-1. Load required skills (mini-plan, todo-management, severity-classification, report-templates)
-2. Create mini-plan for reviewing all 6 pillars
+1. Load required skills (ff-mini-plan, ff-todo-management, ff-severity-classification, ff-report-templates)
+2. Create ff-mini-plan for reviewing all 6 pillars
 3. Create todo list from the plan (one todo per pillar)
 4. Review each pillar, updating todos in real-time
 5. Score each pillar and identify issues
-6. Format output using report-templates (Well-Architected template)
+6. Format output using ff-report-templates (Well-Architected template)
 7. Mark all todos complete before finishing
 8. Invoke other agents if specialized issues found
 

package/agents/planning.md

@@ -157,13 +157,13 @@ Recommend escalation to full architectural planning when:
 
 ## Workflow
 
-1. Load required skills (mini-plan, todo-management, report-templates)
+1. Load required skills (ff-mini-plan, ff-todo-management, ff-report-templates)
 2. Create todo list for planning process
 3. Analyze requirements and mark todo complete
 4. Assess complexity and mark todo complete
 5. Identify files and mark todo complete
 6. Create implementation plan
-7. Format output using report-templates
+7. Format output using ff-report-templates
 8. Mark all todos complete
 9. Hand off plan to @building agent for implementation
 

package/agents/research.md

@@ -19,7 +19,7 @@ You are a research specialist for Feature Factory. Your role is to investigate e
 At the start of EVERY research task:
 
 1. **Check current date** using bash `date` command
-2. **Load the research-methods skill** for research methodology
+2. **Load the ff-research-methods skill** for research methodology
 3. **Load the ff-mini-plan skill** and create a research plan
 4. **Load the ff-todo-management skill** and create a todo list
 5. **Load the ff-report-templates skill** for output formatting
@@ -68,7 +68,7 @@ Use ff-mini-plan skill:
 
 ### Step 3: Execute Research
 
-Follow the research hierarchy from research-methods skill:
+Follow the research hierarchy from ff-research-methods skill:
 
 1. **Web Search** (search_web) - Find current best practices and documentation
 2. **Read URLs** (read_url) - Extract detailed information from specific pages
@@ -279,7 +279,7 @@ Structure research findings using ff-report-templates:
 ## Workflow
 
 1. **Start:** Check current date with bash `date`
-2. **Plan:** Load research-methods, ff-mini-plan skills and create research plan
+2. **Plan:** Load ff-research-methods, ff-mini-plan skills and create research plan
 3. **Track:** Load ff-todo-management and create todo list
 4. **Research:** Execute research using MCP tools following hierarchy
 5. **Synthesize:** Consolidate findings using ff-report-templates
@@ -292,7 +292,7 @@ Structure research findings using ff-report-templates:
 
 ```
 1. bash: date → Sun Feb 1 18:25:00 UTC 2026
-2. Load skills: research-methods, mini-plan, todo-management, report-templates
+2. Load skills: ff-research-methods, ff-mini-plan, ff-todo-management, ff-report-templates
 3. Create plan: Research RSC concepts, docs, examples, patterns
 4. Create todos and track:
    - [x] Check date
@@ -304,7 +304,7 @@ Structure research findings using ff-report-templates:
    - read_url: "https://react.dev/reference/react/use-server"
    - codesearch: "React Server Components examples"
    - gh_grep: "'use server'", language: TypeScript
-6. Format findings with report-templates
+6. Format findings with ff-report-templates
 7. Return structured report to invoking agent
 ```
 

package/agents/reviewing.md

@@ -213,7 +213,7 @@ When combining results from multiple agents:
 
 ## Workflow
 
-1. Load required skills (todo-management, report-templates, severity-classification)
+1. Load required skills (ff-todo-management, ff-report-templates, ff-severity-classification)
 2. Create review todo list
 3. Mark "Launch validation" as in_progress
 4. Invoke @ff-validate agent

package/bin/ff-deploy.js

@@ -33,7 +33,7 @@ const DEFAULT_MCP_CONFIG = {
       type: 'remote',
       url: 'https://mcp.jina.ai/v1',
       headers: {
-        Authorization: 'Bearer jina_5a6cb847e5d44f738658c722c0c25c8bqjnj_wxVri8AuP-3awHMf1Zh6Xi7',
+        Authorization: 'Bearer {env:JINAAI_API_KEY}',
       },
     },
     gh_grep: {

package/package.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json.schemastore.org/package.json",
   "name": "@syntesseraai/opencode-feature-factory",
-  "version": "0.2.1",
+  "version": "0.2.3",
   "description": "OpenCode plugin for Feature Factory agents - provides sub-agents and skills for validation, review, security, and architecture assessment",
   "type": "module",
   "license": "MIT",

package/skills/{research-methods → ff-research-methods}/SKILL.md

@@ -1,5 +1,5 @@
 ---
-name: research-methods
+name: ff-research-methods
 description: Research methodology for investigating external topics using MCP tools. Use this skill to search for current information, best practices, and implementation examples.
 license: MIT
 compatibility: opencode

package/src/index.ts

@@ -17,15 +17,6 @@ export const AGENT_PATHS = {
   'ff-validate': './agents/ff-validate.md',
 };
 
-// Backward compatibility: Export tool plugins that delegate to sub-agents
-// These tools now invoke the corresponding sub-agents instead of returning prompts
-export { FFAcceptancePlugin } from './tools/ffAcceptance';
-export { FFMiniPlanPlugin } from './tools/ffMiniPlan';
-export { FFReviewPlugin } from './tools/ffReview';
-export { FFSecurityPlugin } from './tools/ffSecurity';
-export { FFValidatePlugin } from './tools/ffValidate';
-export { FFWellArchitectedPlugin } from './tools/ffWellArchitected';
-
 const SERVICE_NAME = 'feature-factory';
 
 type Client = PluginInput['client'];

package/src/tools/ffAcceptance.ts

@@ -1,31 +0,0 @@
-import { type Plugin, tool } from '@opencode-ai/plugin';
-
-/**
- * @ff-acceptance Tool Plugin (Backward Compatibility)
- *
- * This tool delegates to the @ff-acceptance sub-agent.
- *
- * NOTE: New code should use the sub-agent directly via @ff-acceptance
- * or invoke it through the @ff-validate orchestrator.
- */
-
-export const FFAcceptancePlugin: Plugin = async (_ctx) => {
-  return {
-    tool: {
-      ff_acceptance: tool({
-        description:
-          'Validates implementation against acceptance criteria (@ff-acceptance), checks if code meets all requirements. DEPRECATED: Use @ff-acceptance sub-agent directly.',
-        args: {
-          task: tool.schema
-            .string()
-            .describe('The validation task or acceptance criteria to check'),
-        },
-        async execute({ task }) {
-          // Invoke the sub-agent via the Task tool
-          // This maintains backward compatibility while using the new sub-agent
-          return `Invoking @ff-acceptance sub-agent to validate: ${task}\n\nPlease wait for the acceptance validation to complete. The @ff-acceptance agent will:\n1. Create a validation plan\n2. Check all acceptance criteria\n3. Provide a detailed report with findings\n\nFor direct access to the sub-agent, use: @ff-acceptance ${task}`;
-        },
-      }),
-    },
-  };
-};

package/src/tools/ffMiniPlan.ts

@@ -1,28 +0,0 @@
-import { type Plugin, tool } from '@opencode-ai/plugin';
-
-/**
- * @ff-mini-plan Tool Plugin (Backward Compatibility)
- *
- * This tool loads the ff-mini-plan skill and provides guidance.
- *
- * NOTE: New code should use the skill directly via the skill tool
- * or invoke it through sub-agents that load it automatically.
- */
-
-export const FFMiniPlanPlugin: Plugin = async (ctx) => {
-  return {
-    tool: {
-      ff_mini_plan: tool({
-        description:
-          'Creates mini implementation plans for smaller tasks (2-5 steps) (@ff-mini-plan), quick planning, small fixes. DEPRECATED: Load the ff-mini-plan skill directly.',
-        args: {
-          task: tool.schema.string().describe('The simple task or issue description to plan for'),
-        },
-        async execute({ task }) {
-          // Provide guidance on using the skill instead
-          return `# Mini Planning\n\nFor task: ${task}\n\n## Using the ff-mini-plan Skill\n\nThe ff-mini-plan skill is now available globally. To use it:\n\n1. **In sub-agents**: Skills are loaded automatically at the start of tasks\n2. **Manually**: Use the skill tool to load ff-mini-plan\n\n## Quick Planning Guide\n\nThe mini-plan skill helps create 2-5 step plans for simple tasks:\n\n### Planning Process:\n1. Assess task complexity (suitable for 2-5 steps?)\n2. Create concrete steps with specific files\n3. Estimate effort and complexity\n4. Identify quick wins\n\n### Escalation Criteria:\n- Requires more than 5 implementation steps\n- Affects multiple components or services\n- Requires architecture decisions\n- Has security implications\n\n### Example Plan Structure:\n\`\`\`\n# Mini Plan: [Task]\n\n**Status:** Ready\n**Time Estimate:** 5-10 minutes\n**Complexity:** Simple\n\n## 📋 Steps\n\n1. **[Step 1]**\n   - What to do: [Action]\n   - Files: \`file.ts\`\n   - Tests: [Verification]\n\n2. **[Step 2]**\n   - What to do: [Action]\n   - Files: \`file.ts\`\n   - Tests: [Verification]\n\n## ⚡ Quick Wins\n\n- [Improvement] - [Benefit]\n\`\`\`\n\nFor comprehensive planning, use @ff-validate to run all validation agents with mini-plans.`;
-        },
-      }),
-    },
-  };
-};

package/src/tools/ffReview.ts

@@ -1,28 +0,0 @@
-import { type Plugin, tool } from '@opencode-ai/plugin';
-
-/**
- * @ff-review Tool Plugin (Backward Compatibility)
- *
- * This tool delegates to the @ff-review sub-agent.
- *
- * NOTE: New code should use the sub-agent directly via @ff-review
- * or invoke it through the @ff-validate orchestrator.
- */
-
-export const FFReviewPlugin: Plugin = async (_ctx) => {
-  return {
-    tool: {
-      ff_review: tool({
-        description:
-          'Reviews code changes for correctness, quality, and test coverage (@ff-review), code review, quality check. DEPRECATED: Use @ff-review sub-agent directly.',
-        args: {
-          task: tool.schema.string().describe('The code review task or diff to analyze'),
-        },
-        async execute({ task }) {
-          // Invoke the sub-agent via the Task tool
-          return `Invoking @ff-review sub-agent to review: ${task}\n\nPlease wait for the code review to complete. The @ff-review agent will:\n1. Create a review plan\n2. Check correctness, quality, and tests\n3. Provide a detailed report with findings\n\nFor direct access to the sub-agent, use: @ff-review ${task}`;
-        },
-      }),
-    },
-  };
-};

package/src/tools/ffSecurity.ts

@@ -1,28 +0,0 @@
-import { type Plugin, tool } from '@opencode-ai/plugin';
-
-/**
- * @ff-security Tool Plugin (Backward Compatibility)
- *
- * This tool delegates to the @ff-security sub-agent.
- *
- * NOTE: New code should use the sub-agent directly via @ff-security
- * or invoke it through the @ff-validate orchestrator.
- */
-
-export const FFSecurityPlugin: Plugin = async (_ctx) => {
-  return {
-    tool: {
-      ff_security: tool({
-        description:
-          'Performs deep security audits on code changes (@ff-security), security check, vulnerability audit. DEPRECATED: Use @ff-security sub-agent directly.',
-        args: {
-          task: tool.schema.string().describe('The security audit task or code to analyze'),
-        },
-        async execute({ task }) {
-          // Invoke the sub-agent via the Task tool
-          return `Invoking @ff-security sub-agent to audit: ${task}\n\nPlease wait for the security audit to complete. The @ff-security agent will:\n1. Create an audit plan\n2. Check for OWASP Top 10 and other vulnerabilities\n3. Provide a detailed security report\n\nFor direct access to the sub-agent, use: @ff-security ${task}`;
-        },
-      }),
-    },
-  };
-};

package/src/tools/ffValidate.ts

@@ -1,28 +0,0 @@
-import { type Plugin, tool } from '@opencode-ai/plugin';
-
-/**
- * @ff-validate Tool Plugin (Backward Compatibility)
- *
- * This tool delegates to the @ff-validate sub-agent.
- *
- * NOTE: New code should use the sub-agent directly via @ff-validate
- * for comprehensive validation.
- */
-
-export const FFValidatePlugin: Plugin = async (_ctx) => {
-  return {
-    tool: {
-      ff_validate: tool({
-        description:
-          'Orchestrates comprehensive validation by running multiple review agents in parallel (@ff-validate), validate changes, comprehensive review. DEPRECATED: Use @ff-validate sub-agent directly.',
-        args: {
-          task: tool.schema.string().describe('The validation task description'),
-        },
-        async execute({ task }) {
-          // Invoke the sub-agent via the Task tool
-          return `Invoking @ff-validate sub-agent to orchestrate comprehensive validation: ${task}\n\nPlease wait for the validation to complete. The @ff-validate agent will:\n1. Create an orchestration plan\n2. Run all validation agents in parallel:\n   - @ff-acceptance (requirements validation)\n   - @ff-review (code quality)\n   - @ff-security (security audit)\n   - @ff-well-architected (architecture review)\n3. Aggregate results and provide a comprehensive report\n\nFor direct access to the orchestrator, use: @ff-validate ${task}`;
-        },
-      }),
-    },
-  };
-};

package/src/tools/ffWellArchitected.ts

@@ -1,30 +0,0 @@
-import { type Plugin, tool } from '@opencode-ai/plugin';
-
-/**
- * @ff-well-architected Tool Plugin (Backward Compatibility)
- *
- * This tool delegates to the @ff-well-architected sub-agent.
- *
- * NOTE: New code should use the sub-agent directly via @ff-well-architected
- * or invoke it through the @ff-validate orchestrator.
- */
-
-export const FFWellArchitectedPlugin: Plugin = async (_ctx) => {
-  return {
-    tool: {
-      ff_well_architected: tool({
-        description:
-          'Reviews code against AWS Well-Architected Framework pillars (@ff-well-architected), architecture review, AWS best practices. DEPRECATED: Use @ff-well-architected sub-agent directly.',
-        args: {
-          task: tool.schema
-            .string()
-            .describe('The architecture review task or system design to analyze'),
-        },
-        async execute({ task }) {
-          // Invoke the sub-agent via the Task tool
-          return `Invoking @ff-well-architected sub-agent to review: ${task}\n\nPlease wait for the architecture review to complete. The @ff-well-architected agent will:\n1. Create a review plan\n2. Assess against AWS Well-Architected 6 pillars:\n   - Operational Excellence\n   - Security\n   - Reliability\n   - Performance Efficiency\n   - Cost Optimization\n   - Sustainability\n3. Provide a detailed report with pillar scores\n\nFor direct access to the sub-agent, use: @ff-well-architected ${task}`;
-        },
-      }),
-    },
-  };
-};