mirrors / wshobson--agents

项目文件夹

wshobson--agents
文件
main
T
Seth Hobson baa5bd7997 feat: add llm-finetuning and dgx-spark-ops plugins (eval-gated fine-tuning lifecycle) (#624) 
* feat(dgx-spark-ops): scaffold plugin and register in marketplace

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(dgx-spark-ops): add spark-environment-setup skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(dgx-spark-ops): tighten spark-environment-setup per review

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(dgx-spark-ops): add spark-training-gotchas skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(dgx-spark-ops): align preflight.sh output contract with docs

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(dgx-spark-ops): add spark-memory-thermal-ops skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(dgx-spark-ops): trim spark-memory-thermal-ops and source 68GB anchor

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(dgx-spark-ops): add dgx-spark-ops-engineer agent

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(dgx-spark-ops): defer agent facts to skills

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(dgx-spark-ops): add /spark-preflight command

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): scaffold plugin and register in marketplace

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add finetuning-method-selection skill with dated model catalog

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add lora-qlora-recipes skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add preference-optimization skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add grpo-rlvr-training skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(llm-finetuning): add isolation warning to execution reward

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add vision-sft skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add dataset-curation skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(llm-finetuning): correct loss-masking example in dataset-curation

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add eval-harness-first skill (Phase 0 gate)

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(llm-finetuning): bring eval-harness-first into line band

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(llm-finetuning): reclaim byte headroom in eval-harness-first

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add trace-to-training-data skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add checkpoint-promotion skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add quantized-export skill

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add llm-finetuning-architect agent

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add llm-finetuning-training-engineer agent

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add llm-finetuning-eval-engineer agent

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add /finetune phase-gated lifecycle command

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(llm-finetuning): thread checkpoint path into /finetune Phase 6

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* feat(llm-finetuning): add /promote-checkpoint command

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(llm-finetuning): thread checkpoint path via phase5.output in /promote-checkpoint

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix(llm-finetuning): robust checkpoint discovery and goldens fingerprint in re-gate

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* docs: register llm-finetuning and dgx-spark-ops (94 plugins, 203 agents, 175 skills, 109 commands)

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* docs(agents): add fine-tuning and spark-ops agent entries (203 agents)

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* chore: generate per-harness artifacts for llm-finetuning and dgx-spark-ops

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix: final-review cleanups for fine-tuning plugins

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix: address PR #624 review feedback (G1 NGC detection, Phase 3 fallback dispatch, registry metadata)

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix: enforce sandbox boundary in execution grader and reward examples

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix: address CodeRabbit review findings on PR #624

Verified and fixed 37 of 43 outstanding CodeRabbit findings across the
llm-finetuning and dgx-spark-ops plugins (skipping 6 confirmed false
positives/already-fixed, with reasons in the disposition report).

Highlights: cross-file contracts (goldens fingerprint persistence,
paired-arena stage numbering, canonical golden-ID field, RERUN
resolution before promotion) now match between finetune.md,
promote-checkpoint.md, and checkpoint-promotion's templates. TRL API
usages (SFTConfig.max_length, trl.experimental ORPO/CPO imports)
verified against live TRL docs rather than blindly renamed. Several
runnable examples hardened against real failure modes: malformed
judge output, empty arena results, non-distinct DPO pairs, unbounded
rejection-sampling fan-out, non-deterministic smoke-test comparisons,
silent FP8-to-bf16 fallback, and orphaned background thermal-sampler
processes. Container detection (G9) and LoRA adapter-size math fixed
in both dgx-spark-ops and llm-finetuning where the same bugs were
independently present in each plugin.

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix: apply dogfood friction-log remediations (F1-F32) from DGX Spark run

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf

* fix: address CodeRabbit round-2 findings (digest pins, suite-size math, monkeypatch scoping)

Claude-Session: https://claude.ai/code/session_01RsN3Vz5fZRTdVMkNtmxhSf
2026-07-14 13:18:02 -04:00

384 行
30 KiB
Markdown
原始文件 永久链接 Blame 文件历史

此文件含有模棱两可的 Unicode 字符
此文件含有可能会与其他字符混淆的 Unicode 字符。 如果您是想特意这样的，可以安全地忽略该警告。 使用 Escape 按钮显示他们。
# Agent Reference
Complete reference for all **203 local specialized AI agents** organized by category with model assignments.
## Agent Categories
### Architecture & System Design
#### Core Architecture
| Agent | Model | Description |
| --------------------------------------------------------------------------------------------- | ------ | ---------------------------------------------------------------------- |
| [backend-architect](../plugins/backend-development/agents/backend-architect.md) | opus | RESTful API design, microservice boundaries, database schemas |
| [frontend-developer](../plugins/multi-platform-apps/agents/frontend-developer.md) | sonnet | React components, responsive layouts, client-side state management |
| [graphql-architect](../plugins/backend-development/agents/graphql-architect.md) | opus | GraphQL schemas, resolvers, federation architecture |
| [architect-reviewer](../plugins/comprehensive-review/agents/architect-review.md) | opus | Architectural consistency analysis and pattern validation |
| [cloud-architect](../plugins/cloud-infrastructure/agents/cloud-architect.md) | opus | AWS/Azure/GCP infrastructure design and cost optimization |
| [hybrid-cloud-architect](../plugins/cloud-infrastructure/agents/hybrid-cloud-architect.md) | opus | Multi-cloud strategies across cloud and on-premises environments |
| [kubernetes-architect](../plugins/kubernetes-operations/agents/kubernetes-architect.md) | opus | Cloud-native infrastructure with Kubernetes and GitOps |
| [service-mesh-expert](../plugins/cloud-infrastructure/agents/service-mesh-expert.md) | opus | Istio/Linkerd service mesh architecture, mTLS, and traffic management |
| [event-sourcing-architect](../plugins/backend-development/agents/event-sourcing-architect.md) | opus | Event sourcing, CQRS patterns, event stores, and saga orchestration |
| [monorepo-architect](../plugins/developer-essentials/agents/monorepo-architect.md) | opus | Monorepo tooling with Nx, Turborepo, Bazel, and workspace optimization |
#### UI/UX & Mobile
| Agent | Model | Description |
| ---------------------------------------------------------------------------------------- | ------ | ------------------------------------------------------- |
| [ui-designer](../plugins/ui-design/agents/ui-designer.md) | opus | UI/UX design for mobile and web with modern patterns |
| [accessibility-expert](../plugins/ui-design/agents/accessibility-expert.md) | opus | WCAG compliance, accessibility audits, inclusive design |
| [design-system-architect](../plugins/ui-design/agents/design-system-architect.md) | opus | Design tokens, component libraries, theming systems |
| [ui-ux-designer](../plugins/multi-platform-apps/agents/ui-ux-designer.md) | sonnet | Interface design, wireframes, design systems |
| [ui-visual-validator](../plugins/accessibility-compliance/agents/ui-visual-validator.md) | sonnet | Visual regression testing and UI verification |
| [mobile-developer](../plugins/multi-platform-apps/agents/mobile-developer.md) | sonnet | React Native and Flutter application development |
| [ios-developer](../plugins/multi-platform-apps/agents/ios-developer.md) | sonnet | Native iOS development with Swift/SwiftUI |
| [flutter-expert](../plugins/multi-platform-apps/agents/flutter-expert.md) | sonnet | Advanced Flutter development with state management |
### Programming Languages
#### Systems & Low-Level
| Agent | Model | Description |
| ----------------------------------------------------------------- | ------ | ----------------------------------------------------------- |
| [c-pro](../plugins/systems-programming/agents/c-pro.md) | sonnet | System programming with memory management and OS interfaces |
| [cpp-pro](../plugins/systems-programming/agents/cpp-pro.md) | sonnet | Modern C++ with RAII, smart pointers, STL algorithms |
| [rust-pro](../plugins/systems-programming/agents/rust-pro.md) | sonnet | Memory-safe systems programming with ownership patterns |
| [golang-pro](../plugins/systems-programming/agents/golang-pro.md) | sonnet | Concurrent programming with goroutines and channels |
#### Web & Application
| Agent | Model | Description |
| ----------------------------------------------------------------------------------- | ------ | --------------------------------------------------------------------------------- |
| [javascript-pro](../plugins/javascript-typescript/agents/javascript-pro.md) | sonnet | Modern JavaScript with ES6+, async patterns, Node.js |
| [typescript-pro](../plugins/javascript-typescript/agents/typescript-pro.md) | sonnet | Advanced TypeScript with type systems and generics |
| [python-pro](../plugins/python-development/agents/python-pro.md) | sonnet | Python development with advanced features and optimization |
| [temporal-python-pro](../plugins/backend-development/agents/temporal-python-pro.md) | sonnet | Temporal workflow orchestration with Python SDK, durable workflows, saga patterns |
| [ruby-pro](../plugins/web-scripting/agents/ruby-pro.md) | sonnet | Ruby with metaprogramming, Rails patterns, gem development |
| [php-pro](../plugins/web-scripting/agents/php-pro.md) | sonnet | Modern PHP with frameworks and performance optimization |
#### Enterprise & JVM
| Agent | Model | Description |
| ----------------------------------------------------------- | ------ | -------------------------------------------------------------------- |
| [java-pro](../plugins/jvm-languages/agents/java-pro.md) | sonnet | Modern Java with streams, concurrency, JVM optimization |
| [scala-pro](../plugins/jvm-languages/agents/scala-pro.md) | sonnet | Enterprise Scala with functional programming and distributed systems |
| [csharp-pro](../plugins/jvm-languages/agents/csharp-pro.md) | sonnet | C# development with .NET frameworks and patterns |
#### Specialized Platforms
| Agent | Model | Description |
| ---------------------------------------------------------------------------------- | ------ | ----------------------------------------------------------------------------------------- |
| [elixir-pro](../plugins/functional-programming/agents/elixir-pro.md) | sonnet | Elixir with OTP patterns and Phoenix frameworks |
| [django-pro](../plugins/api-scaffolding/agents/django-pro.md) | sonnet | Django development with ORM and async views |
| [fastapi-pro](../plugins/api-scaffolding/agents/fastapi-pro.md) | sonnet | FastAPI with async patterns and Pydantic |
| [haskell-pro](../plugins/functional-programming/agents/haskell-pro.md) | sonnet | Strongly typed functional programming with purity, advanced type systems, and concurrency |
| [unity-developer](../plugins/game-development/agents/unity-developer.md) | sonnet | Unity game development and optimization |
| [minecraft-bukkit-pro](../plugins/game-development/agents/minecraft-bukkit-pro.md) | sonnet | Minecraft server plugin development |
| [sql-pro](../plugins/database-design/agents/sql-pro.md) | sonnet | Complex SQL queries and database optimization |
### Infrastructure & Operations
#### DevOps & Deployment
| Agent | Model | Description |
| -------------------------------------------------------------------------------------- | ------ | ------------------------------------------------------------------ |
| [devops-troubleshooter](../plugins/incident-response/agents/devops-troubleshooter.md) | sonnet | Production debugging, log analysis, deployment troubleshooting |
| [deployment-engineer](../plugins/cloud-infrastructure/agents/deployment-engineer.md) | sonnet | CI/CD pipelines, containerization, cloud deployments |
| [terraform-specialist](../plugins/cloud-infrastructure/agents/terraform-specialist.md) | sonnet | Infrastructure as Code with Terraform modules and state management |
| [dx-optimizer](../plugins/team-collaboration/agents/dx-optimizer.md) | sonnet | Developer experience optimization and tooling improvements |
#### Database Management
| Agent | Model | Description |
| -------------------------------------------------------------------------------------- | ------ | ------------------------------------------------------------------- |
| [database-optimizer](../plugins/observability-monitoring/agents/database-optimizer.md) | sonnet | Query optimization, index design, migration strategies |
| [database-admin](../plugins/database-migrations/agents/database-admin.md) | sonnet | Database operations, backup, replication, monitoring |
| [database-architect](../plugins/database-design/agents/database-architect.md) | opus | Database design from scratch, technology selection, schema modeling |
#### Incident Response & Network
| Agent | Model | Description |
| ---------------------------------------------------------------------------------- | ------ | --------------------------------------------------- |
| [incident-responder](../plugins/incident-response/agents/incident-responder.md) | opus | Production incident management and resolution |
| [network-engineer](../plugins/observability-monitoring/agents/network-engineer.md) | sonnet | Network debugging, load balancing, traffic analysis |
#### Project Management
| Agent | Model | Description |
| ----------------------------------------------------------------- | ----- | ------------------------------------------------------------------------------------ |
| [conductor-validator](../plugins/conductor/agents/conductor-validator.md) | opus | Validates Conductor project artifacts for completeness, consistency, and correctness |
### Quality Assurance & Security
#### Code Quality & Review
| Agent | Model | Description |
| ------------------------------------------------------------------------------------------------ | ----- | --------------------------------------------------------------- |
| [code-reviewer](../plugins/comprehensive-review/agents/code-reviewer.md) | opus | Code review with security focus and production reliability |
| [security-auditor](../plugins/comprehensive-review/agents/security-auditor.md) | opus | Vulnerability assessment and OWASP compliance |
| [backend-security-coder](../plugins/data-validation-suite/agents/backend-security-coder.md) | opus | Secure backend coding practices, API security implementation |
| [frontend-security-coder](../plugins/frontend-mobile-security/agents/frontend-security-coder.md) | opus | XSS prevention, CSP implementation, client-side security |
| [mobile-security-coder](../plugins/frontend-mobile-security/agents/mobile-security-coder.md) | opus | Mobile security patterns, WebView security, biometric auth |
| [threat-modeling-expert](../plugins/security-scanning/agents/threat-modeling-expert.md) | opus | STRIDE threat modeling, attack trees, and security requirements |
#### Testing & Debugging
| Agent | Model | Description |
| ----------------------------------------------------------------------------- | ------ | ---------------------------------------------------------- |
| [test-automator](../plugins/codebase-cleanup/agents/test-automator.md) | sonnet | Comprehensive test suite creation (unit, integration, e2e) |
| [tdd-orchestrator](../plugins/backend-development/agents/tdd-orchestrator.md) | sonnet | Test-Driven Development methodology guidance |
| [debugger](../plugins/error-debugging/agents/debugger.md) | sonnet | Error resolution and test failure analysis |
| [error-detective](../plugins/error-debugging/agents/error-detective.md) | sonnet | Log analysis and error pattern recognition |
#### Performance & Observability
| Agent | Model | Description |
| ---------------------------------------------------------------------------------------------- | ----- | -------------------------------------------------------------- |
| [performance-engineer](../plugins/observability-monitoring/agents/performance-engineer.md) | opus | Application profiling and optimization |
| [observability-engineer](../plugins/observability-monitoring/agents/observability-engineer.md) | opus | Production monitoring, distributed tracing, SLI/SLO management |
| [search-specialist](../plugins/content-marketing/agents/search-specialist.md) | haiku | Advanced web research and information synthesis |
### Data & AI
#### Data Engineering & Analytics
| Agent | Model | Description |
| -------------------------------------------------------------------------- | ------ | ------------------------------------------------------- |
| [data-scientist](../plugins/machine-learning-ops/agents/data-scientist.md) | opus | Data analysis, SQL queries, BigQuery operations |
| [data-engineer](../plugins/data-engineering/agents/data-engineer.md) | sonnet | ETL pipelines, data warehouses, streaming architectures |
#### Machine Learning & AI
| Agent | Model | Description |
| --------------------------------------------------------------------------------------------- | ----- | --------------------------------------------------------------------- |
| [ai-engineer](../plugins/llm-application-dev/agents/ai-engineer.md) | opus | LLM applications, RAG systems, prompt pipelines |
| [ml-engineer](../plugins/machine-learning-ops/agents/ml-engineer.md) | opus | ML pipelines, model serving, feature engineering |
| [mlops-engineer](../plugins/machine-learning-ops/agents/mlops-engineer.md) | opus | ML infrastructure, experiment tracking, model registries |
| [prompt-engineer](../plugins/llm-application-dev/agents/prompt-engineer.md) | opus | LLM prompt optimization and engineering |
| [vector-database-engineer](../plugins/llm-application-dev/agents/vector-database-engineer.md) | opus | Vector databases, embeddings, similarity search, and hybrid retrieval |
| [llm-finetuning-architect](../plugins/llm-finetuning/agents/llm-finetuning-architect.md) | opus | Fine-tuning strategy, method/model selection, eval-gate ownership |
| [llm-finetuning-training-engineer](../plugins/llm-finetuning/agents/llm-finetuning-training-engineer.md) | sonnet | Dataset prep, Unsloth training runs, artifact export |
| [llm-finetuning-eval-engineer](../plugins/llm-finetuning/agents/llm-finetuning-eval-engineer.md) | sonnet | Golden sets, judge calibration, checkpoint promotion verdicts |
| [dgx-spark-ops-engineer](../plugins/dgx-spark-ops/agents/dgx-spark-ops-engineer.md) | sonnet | DGX Spark (GB10/aarch64/CUDA 13) environment setup and diagnostics |
### Documentation & Technical Writing
| Agent | Model | Description |
| ------------------------------------------------------------------------------------ | ------ | --------------------------------------------------------------------- |
| [docs-architect](../plugins/code-documentation/agents/docs-architect.md) | opus | Comprehensive technical documentation generation |
| [api-documenter](../plugins/api-testing-observability/agents/api-documenter.md) | sonnet | OpenAPI/Swagger specifications and developer docs |
| [reference-builder](../plugins/documentation-generation/agents/reference-builder.md) | haiku | Technical references and API documentation |
| [tutorial-engineer](../plugins/code-documentation/agents/tutorial-engineer.md) | sonnet | Step-by-step tutorials and educational content |
| [mermaid-expert](../plugins/documentation-generation/agents/mermaid-expert.md) | sonnet | Diagram creation (flowcharts, sequences, ERDs) |
| [c4-code](../plugins/c4-architecture/agents/c4-code.md) | haiku | C4 Code-level documentation with function signatures and dependencies |
| [c4-component](../plugins/c4-architecture/agents/c4-component.md) | sonnet | C4 Component-level architecture synthesis and documentation |
| [c4-container](../plugins/c4-architecture/agents/c4-container.md) | sonnet | C4 Container-level architecture with API documentation |
| [c4-context](../plugins/c4-architecture/agents/c4-context.md) | sonnet | C4 Context-level system documentation with personas and user journeys |
### Business & Operations
#### Business Analysis & Finance
| Agent | Model | Description |
| ---------------------------------------------------------------------------- | ------ | ------------------------------------------------------- |
| [business-analyst](../plugins/business-analytics/agents/business-analyst.md) | sonnet | Metrics analysis, reporting, KPI tracking |
| [quant-analyst](../plugins/quantitative-trading/agents/quant-analyst.md) | opus | Financial modeling, trading strategies, market analysis |
| [risk-manager](../plugins/quantitative-trading/agents/risk-manager.md) | sonnet | Portfolio risk monitoring and management |
#### Marketing & Sales
| Agent | Model | Description |
| --------------------------------------------------------------------------------- | ------ | -------------------------------------------- |
| [content-marketer](../plugins/content-marketing/agents/content-marketer.md) | sonnet | Blog posts, social media, email campaigns |
| [social-publishing-publisher](../plugins/social-publishing/agents/social-publishing-publisher.md) | haiku | Multi-platform social media publishing via SocialClaw |
| [sales-automator](../plugins/customer-sales-automation/agents/sales-automator.md) | haiku | Cold emails, follow-ups, proposal generation |
#### Support & Legal
| Agent | Model | Description |
| ----------------------------------------------------------------------------------- | ------ | ------------------------------------------------------- |
| [customer-support](../plugins/customer-sales-automation/agents/customer-support.md) | sonnet | Support tickets, FAQ responses, customer communication |
| [hr-pro](../plugins/hr-legal-compliance/agents/hr-pro.md) | opus | HR operations, policies, employee relations |
| [legal-advisor](../plugins/hr-legal-compliance/agents/legal-advisor.md) | opus | Privacy policies, terms of service, legal documentation |
### SEO & Content Optimization
| Agent | Model | Description |
| --------------------------------------------------------------------------------------------------------- | ------ | ---------------------------------------------------- |
| [seo-content-auditor](../plugins/seo-content-creation/agents/seo-content-auditor.md) | sonnet | Content quality analysis, E-E-A-T signals assessment |
| [seo-meta-optimizer](../plugins/seo-technical-optimization/agents/seo-meta-optimizer.md) | haiku | Meta title and description optimization |
| [seo-keyword-strategist](../plugins/seo-technical-optimization/agents/seo-keyword-strategist.md) | haiku | Keyword analysis and semantic variations |
| [seo-structure-architect](../plugins/seo-technical-optimization/agents/seo-structure-architect.md) | haiku | Content structure and schema markup |
| [seo-snippet-hunter](../plugins/seo-technical-optimization/agents/seo-snippet-hunter.md) | haiku | Featured snippet formatting |
| [seo-content-refresher](../plugins/seo-analysis-monitoring/agents/seo-content-refresher.md) | haiku | Content freshness analysis |
| [seo-cannibalization-detector](../plugins/seo-analysis-monitoring/agents/seo-cannibalization-detector.md) | haiku | Keyword overlap detection |
| [seo-authority-builder](../plugins/seo-analysis-monitoring/agents/seo-authority-builder.md) | sonnet | E-E-A-T signal analysis |
| [seo-content-writer](../plugins/seo-content-creation/agents/seo-content-writer.md) | sonnet | SEO-optimized content creation |
| [seo-content-planner](../plugins/seo-content-creation/agents/seo-content-planner.md) | haiku | Content planning and topic clusters |
### Specialized Domains
| Agent | Model | Description |
| --------------------------------------------------------------------------------------- | ------ | ------------------------------------------------------- |
| [arm-cortex-expert](../plugins/arm-cortex-microcontrollers/agents/arm-cortex-expert.md) | sonnet | ARM Cortex-M firmware and peripheral driver development |
| [blockchain-developer](../plugins/blockchain-web3/agents/blockchain-developer.md) | sonnet | Web3 apps, smart contracts, DeFi protocols |
| [payment-integration](../plugins/payment-processing/agents/payment-integration.md) | sonnet | Payment processor integration (Stripe, PayPal) |
| [legacy-modernizer](../plugins/framework-migration/agents/legacy-modernizer.md) | sonnet | Legacy code refactoring and modernization |
| [context-manager](../plugins/agent-orchestration/agents/context-manager.md) | haiku | Multi-agent context management |
## Model Configuration
Agents are assigned to specific Claude models based on task complexity and computational requirements.
### Model Distribution Summary
| Model | Agent Count | Use Case |
| ------- | ----------- | --------------------------------------------------------------- |
| Fable | 0 | Longest-horizon autonomous work (tier above Opus; see criteria) |
| Opus | 55 | Critical architecture, security, code review, production coding |
| Sonnet | 71 | Complex tasks, support with intelligence |
| Haiku | 25 | Fast operational tasks |
| Inherit | 52 | Complex tasks where the user chooses the model at runtime |
### Model Selection Criteria
#### Haiku - Fast Execution & Deterministic Tasks
**Use when:**
- Generating code from well-defined specifications
- Creating tests following established patterns
- Writing documentation with clear templates
- Executing infrastructure operations
- Performing database query optimization
- Handling customer support responses
- Processing SEO optimization tasks
- Managing deployment pipelines
#### Sonnet - Complex Reasoning & Architecture
**Use when:**
- Designing system architecture
- Making technology selection decisions
- Performing security audits
- Reviewing code for architectural patterns
- Creating complex AI/ML pipelines
- Providing language-specific expertise
- Orchestrating multi-agent workflows
- Handling business-critical legal/HR matters
#### Fable - Longest-Horizon Autonomous Work
Claude Fable 5 (`model: fable`) is the tier above Opus. It is opt-in in Claude Code
(v2.1.170+, never the default) and carries roughly 2.6× the effective cost of Opus
($10/$50 per MTok plus a ~30% heavier tokenizer), so reserve it for agents where Opus
demonstrably needs multiple attempts.
**Use when:**
- Running multi-hour autonomous sessions (large codebase migrations, overnight refactors)
- Executing end-to-end work from a single well-specified goal
- Coordinating long-lived parallel sub-agent workstreams
**Avoid for:**
- Security-analysis agents — Fable 5's cyber/bio safety classifiers fall back to
Opus 4.8 on that content anyway
- Anything Sonnet or Opus already completes in one pass
Other harnesses map `fable` to their top available model (see
[authoring.md](authoring.md#model-aliases)). Fable-tier agent bodies should state goals
and constraints rather than step-by-step scaffolding, and must not ask the model to echo
its internal reasoning.
### Hybrid Orchestration Patterns
The plugin ecosystem leverages Sonnet + Haiku orchestration for optimal performance and cost efficiency:
#### Pattern 1: Planning → Execution
```
Sonnet: backend-architect (design API architecture)
↓
Haiku: Generate API endpoints following spec
↓
Haiku: test-automator (generate comprehensive tests)
↓
Sonnet: code-reviewer (architectural review)
```
#### Pattern 2: Reasoning → Action (Incident Response)
```
Sonnet: incident-responder (diagnose issue, create strategy)
↓
Haiku: devops-troubleshooter (execute fixes)
↓
Haiku: deployment-engineer (deploy hotfix)
↓
Haiku: Implement monitoring alerts
```
#### Pattern 3: Complex → Simple (Database Design)
```
Sonnet: database-architect (schema design, technology selection)
↓
Haiku: sql-pro (generate migration scripts)
↓
Haiku: database-admin (execute migrations)
↓
Haiku: database-optimizer (tune query performance)
```
#### Pattern 4: Multi-Agent Workflows
```
Full-Stack Feature Development:
Sonnet: backend-architect + frontend-developer (design components)
↓
Haiku: Generate code following designs
↓
Haiku: test-automator (unit + integration tests)
↓
Sonnet: security-auditor (security review)
↓
Haiku: deployment-engineer (CI/CD setup)
↓
Haiku: Setup observability stack
```
## Agent Invocation
### Natural Language
Agents can be invoked through natural language when you need Claude to reason about which specialist to use:
```
"Use backend-architect to design the authentication API"
"Have security-auditor scan for OWASP vulnerabilities"
"Get performance-engineer to optimize this database query"
```
### Slash Commands
Many agents are accessible through plugin slash commands for direct invocation:
```bash
/backend-development:feature-development user authentication
/security-scanning:security-sast
/incident-response:smart-fix "memory leak in payment service"
```
## Contributing
To add a new agent:
1. Create `plugins/{plugin-name}/agents/{agent-name}.md`
2. Add frontmatter with name, description, and model assignment
3. Write comprehensive system prompt
4. Update plugin definition in `.claude-plugin/marketplace.json`
See [Contributing Guide](../CONTRIBUTING.md) for details.
在新工单中引用 查看 Git Blame 复制永久链接