UNPKG

aios-core

Version:

Synkra AIOS: AI-Orchestrated System for Full Stack Development - Core Framework

341 lines (231 loc) 8.29 kB
# {Agent ID} - {Checklist Title} **Agent:** {AgentName} ({Archetype}) **Purpose:** {Brief description of what this checklist validates} **When to Use:** {When this checklist should be executed} --- ## Overview {1-2 sentence explanation of checklist purpose and scope} **Completion Criteria:** All checkboxes marked with [x] before proceeding. --- ## Pre-Execution Checks **Purpose:** Validate prerequisites before starting main workflow - [ ] {Standard prerequisite check 1 - same across all agents} - [ ] {Standard prerequisite check 2 - same across all agents} - [ ] {Standard prerequisite check 3 - same across all agents} - [ ] {Agent-specific prerequisite if needed} --- ## Execution Validation **Purpose:** Verify correct execution during workflow ### Phase 1: {Phase Name} - [ ] {Validation item 1} - [ ] {Validation item 2} - [ ] {Validation item 3} ### Phase 2: {Phase Name} - [ ] {Validation item 4} - [ ] {Validation item 5} - [ ] {Validation item 6} --- ## Quality Checks **Purpose:** Ensure output meets quality standards ### Code Quality (if applicable) - [ ] All tests passing ({tests.total} tests) - [ ] Code coverage {threshold}% (current: {coverage}%) - [ ] Linting clean (0 errors, 0 warnings) - [ ] No console.log or debug statements - [ ] No commented-out code blocks ### Documentation Quality - [ ] All public functions documented - [ ] README updated if needed - [ ] Inline comments for complex logic - [ ] Change log entry added ### Security & Performance - [ ] No secrets in code - [ ] No SQL injection vulnerabilities - [ ] No XSS vulnerabilities - [ ] Performance benchmarks met - [ ] Memory leaks checked --- ## Post-Execution Review **Purpose:** Final validation before completion - [ ] {Standard review item 1} - [ ] {Standard review item 2} - [ ] {Standard review item 3} - [ ] Output follows standardized template - [ ] Duration and token metrics recorded - [ ] All dependencies satisfied --- ## Metrics Validation **Required metrics tracked:** - [ ] **Duration:** {duration} recorded - [ ] **Tokens Used:** {tokens.total} recorded - [ ] **Tests:** {tests.passed}/{tests.total} ratio calculated - [ ] **Coverage:** {coverage}% measured - [ ] **Linting:** Status recorded **Task-specific metrics (if applicable):** - [ ] {Metric 1}: {value} recorded - [ ] {Metric 2}: {value} recorded --- ## Agent-Specific Guidance **{AgentName} ({Archetype}) Note:** {PERSONALITY_SLOT: Personalized guidance based on archetype} **Examples by archetype:** ### Builder (Dex) > "Se algum teste falhou, refatore até passar. Não entregue código quebrado. Construa com qualidade desde o início." ### Guardian (Quinn) > "Valide edge cases extras além dos listados. Proteção nunca é demais. Cada checklist é uma linha de defesa." ### Balancer (Pax) > "Equilibre velocidade com qualidade. Se precisar escolher, priorize o que agrega mais valor ao usuário final." ### Visionary (Morgan) > "Pense 3 sprints à frente. Cada decisão agora impacta a arquitetura futura. Planeje pensando em escala." ### Flow Master (River) > "Adapte a checklist ao contexto. Se algo não faz sentido neste caso específico, documente e ajuste o processo." --- ## Failure Protocols ### If Pre-Execution Checks Fail **Action:** HALT workflow immediately **Steps:** 1. Document which check failed 2. Notify user with {PERSONALITY_SLOT: personalized error message} 3. Wait for resolution before proceeding **Personalized error messages:** - **Builder:** "⚠️ Pré-requisitos faltando. Não posso construir sem fundação sólida." - **Guardian:** "⚠️ Pré-condições não satisfeitas. Bloqueando execução por segurança." - **Balancer:** "⚠️ Falta consenso nos pré-requisitos. Vamos alinhar antes de prosseguir." ### If Execution Validation Fails **Action:** Roll back to last stable state **Steps:** 1. Identify failed validation item 2. Determine root cause 3. Fix issue 4. Re-run validation 5. Continue only when all items pass ### If Quality Checks Fail **Action:** Refactor until standards met **Steps:** 1. Prioritize CRITICAL issues (security, breaking changes) 2. Address HIGH issues (performance, major bugs) 3. Document MEDIUM/LOW issues for backlog 4. Re-validate after fixes --- ## Reporting Template After completing checklist, generate report using standardized format: ```markdown ## 📊 Checklist Execution Report **Agent:** {agent.name} ({agent.persona_profile.archetype}) **Checklist:** {checklist.name} **Started:** {timestamp.start} **Completed:** {timestamp.end} **Duration:** {duration} **Tokens Used:** {tokens.total} total --- ### Status {PERSONALITY_SLOT: status_message} **Examples:** - "Checklist completa! Todos os 24 itens validados." (Builder) - "Validação rigorosa concluída. Zero pendências." (Guardian) - "Checklist equilibrada. Priorizei itens críticos." (Balancer) ### Results - **Total Items:** {total_items} - **Completed:** {completed_items} - **Failed:** {failed_items} - **Skipped:** {skipped_items} ⚠️ ### Failed Items (if any) {list_of_failed_items_with_reasons} ### Metrics - Duration: {duration} - Tokens: {tokens.total} total - Pass Rate: {pass_rate}% --- {agent.signature_closing} ``` --- ## Integration with Tasks **How tasks use this checklist:** 1. Task loads checklist via dependency 2. Task executes pre-execution checks 3. Task runs main workflow with validation checks 4. Task performs quality checks 5. Task completes post-execution review 6. Task generates checklist report **Example task integration:** ```markdown # develop-story.md (excerpt) ## Completion Workflow 1. All tasks and subtasks marked [x] 2. All tests pass 3. Execute `.aios-core/product/checklists/story-dod-checklist.md` 4. Generate checklist report 5. Set story status: "Ready for Review" ``` --- ## Automation Hooks ### Pre-Commit Hook ```bash #!/bin/bash # .git/hooks/pre-commit echo "🔍 Running {checklist-name}..." # Execute checklist validation node .aios-core/scripts/validate-checklist.js {checklist-id} if [ $? -ne 0 ]; then echo "❌ Checklist validation failed. Commit blocked." exit 1 fi echo "✅ Checklist passed. Proceeding with commit." ``` ### CI/CD Integration ```yaml # .github/workflows/quality-gate.yml jobs: checklist-validation: runs-on: ubuntu-latest steps: - name: Run {Checklist Name} run: | node .aios-core/scripts/validate-checklist.js {checklist-id} - name: Generate Report run: | node .aios-core/scripts/generate-checklist-report.js {checklist-id} - name: Upload Artifacts uses: actions/upload-artifact@v2 with: name: checklist-report path: reports/{checklist-id}-report.md ``` --- ## Continuous Improvement ### Monthly Review - [ ] Review failed items from last 30 days - [ ] Identify patterns in failures - [ ] Update checklist to prevent common issues - [ ] Add new items based on lessons learned ### Quarterly Audit - [ ] Validate all items still relevant - [ ] Remove obsolete checks - [ ] Benchmark against industry standards - [ ] Collect team feedback on usability --- ## Related Documents - **Standard:** `.aios-core/docs/standards/AGENT-PERSONALIZATION-STANDARD-V1.md` - **Agent File:** `.aios-core/agents/{agent-id}.md` - **Tasks Using This:** {list of tasks that execute this checklist} --- ## Validation **Before committing this checklist:** - [ ] All sections have at least 3 items - [ ] Agent-specific guidance matches archetype - [ ] Personalized messages use agent vocabulary - [ ] Reporting template follows standard structure - [ ] Metrics validation section complete - [ ] Failure protocols defined --- **Checklist Version:** 1.0 **Last Updated:** 2025-01-14 **Next Review:** {Date} **Applies to:** Story 6.1.2+ (Personalized Agents) --- **{AgentName} Final Note:** {PERSONALITY_SLOT: Closing message that reinforces archetype values} **Examples:** - **Dex (Builder):** "Checklists são blueprints. Siga-os à risca e construa com confiança. 🔨" - **Quinn (Guardian):** "Cada item é uma proteção. Não pule nenhum. Sua segurança depende disso. 🛡️" - **Pax (Balancer):** "Use bom senso. Checklist é guia, não prisão. Adapte ao contexto. ⚖️"