ai-continuous-training-maintainer
Manter o pipeline de treino contínuo de IA em `E:/GitHub/aidiabetic-research/src/` sempre: 1. Atualizado com **melhores práticas** de engenharia de ML (SOLID, OWASP, OWASP AI) 2. **Livre de CVEs conhecidas** (rastreamento de múltiplas fontes: NVD, GitHub Advisories, OSV, PyPA, etc.) 3. **Auditado por SAST/Trivy** com correção de achados críticos e altos 4. Com **dependências atualizadas** sem quebrar compatibilidade 5. **Sem regressões** — alterações graduais, testadas e com rollback 6. Com **co Use when the user requests ai continuous training maintainer or invokes ai-continuous-training-maintainer. Do NOT use for unrelated languages, domains or production deployment.
- License 许可
- CC-BY-4.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- ai-continuous-training-maintainer
- license
- CC-BY-4.0
- metadata
- { "author": "AEOS", "version": "1.0.0", "owner_agent": "codenavi-agent" }
- description
- Manter o pipeline de treino contínuo de IA em `E:/GitHub/aidiabetic-research/src/` sempre: 1. Atualizado com **melhores práticas** de engenharia de ML (SOLID, OWASP, OWASP AI) 2. **Livre de CVEs conhecidas** (rastreamento de múltiplas fontes: NVD, GitHub Advisories, OSV, PyPA, etc.) 3. **Auditado por SAST/Trivy** com correção de achados críticos e altos 4. Com **dependências atualizadas** sem quebrar compatibilidade 5. **Sem regressões** — alterações graduais, testadas e com rollback 6. Com **co Use when the user requests ai continuous training maintainer or invokes ai-continuous-training-maintainer. Do NOT use for unrelated languages, domains or production deployment.
02 资源文件 50
- 参考文档CONTINUOUS_LEARNING.md1.6 KB
- 参考文档HANDOFF.md2.4 KB
- 参考文档KNOWLEDGE_PROMOTION.md1.8 KB
- 参考文档knowledge/CONTINUOUS_LEARNING.md904 B
- 参考文档knowledge/KNOWLEDGE_PROMOTION.md948 B
- 参考文档knowledge/KNOWLEDGE.md1001 B
- 参考文档knowledge/NEGATIVE_KNOWLEDGE.md2.1 KB
- 参考文档knowledge/POSITIVE_KNOWLEDGE.md1.8 KB
- 参考文档MANIFEST.json4.6 KB
- 参考文档MEMORY_SCHEMA.md2.2 KB
- 参考文档memory/EXECUTIONS.md200 B
- 参考文档memory/FAILURES.md179 B
- 参考文档memory/LESSONS.md142 B
- 参考文档memory/PATTERNS.md151 B
- 参考文档references/instructions/01-purpose-and-contract.md3.0 KB
- 参考文档references/instructions/02-1-identity.md962 B
- 参考文档references/instructions/03-2-mission.md776 B
- 参考文档references/instructions/04-3-activation.md647 B
- 参考文档references/instructions/05-4-non-activation.md473 B
- 参考文档references/instructions/06-5-scope.md1015 B
- 参考文档references/instructions/07-6-authority.md769 B
- 参考文档references/instructions/08-7-inputs.md465 B
- 参考文档references/instructions/09-8-outputs.md883 B
- 参考文档references/instructions/10-9-preconditions.md472 B
- 参考文档references/instructions/11-10-workflow-mandatory-execution-sequence.md12.6 KB
- 参考文档references/instructions/12-11-tool-policy.md723 B
- 参考文档references/instructions/13-12-evidence-policy.md570 B
- 参考文档references/instructions/14-13-subagent-policy.md939 B
- 参考文档references/instructions/15-14-validation.md617 B
- 参考文档references/instructions/16-15-stop-conditions.md1.1 KB
- 参考文档references/instructions/17-16-failure-behavior.md563 B
- 参考文档references/instructions/18-17-completion-criteria.md1.2 KB
- 参考文档references/instructions/19-18-memory-behavior.md971 B
- 参考文档references/instructions/20-19-security-restrictions.md561 B
- 参考文档references/instructions/21-20-cve-sources-multi-source.md768 B
- 参考文档references/instructions/22-21-examples.md2.6 KB
- 参考文档references/instructions/23-22-version.md365 B
- 参考文档references/instructions/24-23-mcp-integration.md854 B
- 参考文档references/instructions/INDEX.md2.1 KB
- 参考文档references/lenses/retained/legacy-contract.md4.0 KB
- 参考文档references/package-notes.md2.2 KB
- 参考文档references/provenance.md579 B
- 参考文档schemas/execution.schema.json1.5 KB
- 参考文档schemas/finding.schema.json1.1 KB
- 参考文档schemas/score.schema.json2.7 KB
- 参考文档templates/cve_report.md990 B
- 参考文档templates/report.md1012 B
- 参考文档tests/CONTINUOUS_LEARNING.md369 B
- 参考文档tests/KNOWLEDGE_PROMOTION.md361 B
- 参考文档tests/POSITIVE_KNOWLEDGE.md466 B
03 SKILL.md 正文
AI Continuous Training Maintainer
Governance: CodENavi Full Workspace v2. Skill of codenavi-agent; inherit AGENTS.md.
Lifecycle: BRIEFING → RECON → PLAN → EXECUTE → VERIFY → DEBRIEF.
Source roles/workers are internal lenses. Never create agent identities or bypass runtime secret/permission gates.
Mission
Manter o pipeline de treino contínuo de IA em E:/GitHub/aidiabetic-research/src/ sempre: 1. Atualizado com melhores práticas de engenharia de ML (SOLID, OWASP, OWASP AI) 2. Livre de CVEs conhecidas (rastreamento de múltiplas fontes: NVD, GitHub Advisories, OSV, PyPA, etc.) 3. Auditado por SAST/Trivy com correção de achados críticos e altos 4. Com dependências atualizadas sem quebrar compatibilidade 5. Sem regressões — alterações graduais, testadas e com rollback 6. Com **co
Workflow and module routing
Read the contract/input/security modules before tools; then follow the applicable procedure in the original order. Load optional modes only when their scope applies.
-
- Purpose and contract — before acting on this contract.
-
- 1. Identity — before acting on this contract.
-
- 2. Mission — before acting on this contract.
-
- 3. Activation — when this domain or mode is in scope.
-
- 4. Non-activation — when this domain or mode is in scope.
-
- 5. Scope — before acting on this contract.
-
- 6. Authority — before acting on this contract.
-
- 7. Inputs — before acting on this contract.
-
- 8. Outputs — before acting on this contract.
-
- 9. Preconditions — before acting on this contract.
-
- 10. Workflow — Mandatory Execution Sequence — at this step of execution.
-
- 11. Tool Policy — before acting on this contract.
-
- 12. Evidence Policy — before acting on this contract.
-
- 13. Subagent Policy — before acting on this contract.
-
- 14. Validation — before claiming completion.
-
- 15. Stop Conditions — when an input, tool or check fails.
-
- 16. Failure Behavior — when an input, tool or check fails.
-
- 17. Completion Criteria — before claiming completion.
-
- 18. Memory Behavior — when reading or promoting continuity knowledge.
-
- 19. Security Restrictions — before acting on this contract.
-
- 20. CVE Sources (Multi-Source) — when this domain or mode is in scope.
-
- 21. Examples — when choosing an output or matching a use case.
-
- 22. Version — when this domain or mode is in scope.
-
- 23. MCP Integration — when this domain or mode is in scope.
Boundaries and result
Use only tools actually available for the authorized task. Keep secrets in runtime providers. Missing evidence, inputs or permissions blocks the affected action. Return the domain output with source pointers, executed checks and remaining uncertainty; never claim an unexecuted check passed.
Provenance and source freshness: provenance.
Read package-notes.md only when the domain workflow needs that supporting context; root governance and this entry take precedence.
04 同仓库更多 Skills
workspace-brain
Operate the CodENavi workspace brain MCP and CLI — cloud DeepSeek RAG, disk corpus at WORKSPACE_BRAIN_ROOT, Jev URL curation, and event logging for Cursor/Codex/OpenCode/Claude. Use when ingesting repo knowledge, asking the brain, mapping web seeds, or recording AI-driven file changes. Do NOT use for local Ollama RAG or storing API keys in git.
workbench-profile-resolver
Provide enterprise-grade capability for `workbench-profile-resolver` inside AEOS v1.1. Use when the user requests workbench profile resolver or invokes workbench-profile-resolver. Do NOT use for unrelated tools or unapproved remote writes.
wordpress-expert
Staff-level WordPress engineering super-skill for one-shot site mapping, clean-code AI-assisted project creation/modification, the complete installed plugin universe, governed remote wp-admin operations, themes/blocks/layout, third-party integrations and production-safe changes. Use when the user requests wordpress expert or invokes wordpress-expert. Do NOT use for unrelated languages, domains or production deployment.
whatsapp-simulator
Generate synthetic WhatsApp-style conversation fixtures for testing conversational systems, support flows and automation. Use when the user requests whatsapp simulator or invokes whatsapp-simulator. Do NOT use for unrelated tools or unapproved remote writes.
web-quality-audit
Assimilated TLC/web-quality multi-area audit. Review performance, accessibility, SEO and best practices in one pass. Triggers: audit my site, review web quality, check page quality, optimize my website across areas. Do not use for a single-area job (core-web-vitals, ai-seo, best-practices, perf-lighthouse).
voiceai
Provide an evidence-backed voice interface for AEOS. Activating this skill is treated as explicit current-session consent for `voiceai` to record user audio or transcribe audio from user-authorized audio/video media within the declared scope. The skill can: Use when the user requests voiceai or invokes voiceai. Do NOT use for unrelated tools or unapproved remote writes.