bootspring_agent

Invoke specialized expert agents for domain-specific guidance and implementation help.

Overview#

bootspring_agent loads a specialized agent profile and provides expert-level guidance tailored to that domain. Each agent has deep knowledge in specific areas like frontend development, database design, security, or business strategy.

Parameters#

ParameterTypeRequiredDescription
agentstringYesAgent name (e.g., "frontend-expert")
taskstringYesThe task or question for the agent
contextstringNoAdditional context
outputFormatstringNo"guidance", "code", "review", "plan"
includeExamplesbooleanNoInclude code examples (default: true)
verbositystringNo"brief", "normal", "detailed"

Usage Examples#

Basic Invocation#

Use the bootspring_agent tool with agent="frontend-expert" and task="Create a responsive sidebar navigation"

With Context#

Use bootspring_agent with agent="database-expert", task="Optimize this query", and context="Query takes 3 seconds to return 1000 users with their orders"

Specific Output Format#

Use bootspring_agent with agent="security-expert", task="Review the authentication flow", and outputFormat="review"

Code-Only Output#

Use bootspring_agent with agent="backend-expert", task="Create a rate limiter middleware", and outputFormat="code"

Available Agents#

Technical Agents (Free Tier)#

AgentExpertise
frontend-expertReact, Next.js, Vue, CSS, accessibility
backend-expertNode.js, APIs, server architecture
database-expertSQL, Prisma, MongoDB, schema design
api-expertREST, GraphQL, tRPC, OpenAPI
testing-expertVitest, Jest, Playwright, TDD
security-expertOWASP, auth, encryption, auditing
performance-expertCaching, optimization, profiling
devops-expertCI/CD, Docker, Kubernetes
ui-ux-expertDesign systems, accessibility, UX
architecture-expertSystem design, patterns, scalability
code-review-expertBest practices, refactoring, quality
vercel-expertVercel deployment, Edge functions

Platform Agents (Free Tier)#

AgentExpertise
auth-expertJWT, Auth.js, Auth0, sessions
payment-expertStripe, subscriptions, billing
email-expertResend, SendGrid, templates
ai-integration-expertLLMs, embeddings, RAG
monitoring-expertSentry, logging, APM
data-modeling-expertERD, normalization, schemas
railway-expertRailway deployment
mobile-expertReact Native, Flutter
infrastructure-expertAWS, GCP, Azure

Business Agents (Pro Tier)#

AgentExpertise
business-strategy-expertStrategy, positioning, GTM
growth-expertPLG, retention, metrics
financial-expertModeling, projections
fundraising-expertPitch decks, investors
legal-expertContracts, compliance
marketing-expertSEO, content, campaigns
sales-expertB2B sales, negotiations
product-expertRoadmapping, prioritization
operations-expertProcess, scaling
competitive-analysis-expertMarket research
content-expertTechnical writing
research-expertTechnology evaluation

Enterprise Agents (Team Tier)#

AgentExpertise
investor-relations-expertBoard management
partnerships-expertStrategic partnerships
private-equity-expertM&A, due diligence

Response Format#

Loading code block...

Examples#

Frontend Expert#

Request:

Use bootspring_agent with agent="frontend-expert" and task="Create an accessible dropdown menu with keyboard navigation"

Response includes:

  • Complete React component with TypeScript
  • Keyboard navigation (arrow keys, escape, enter)
  • ARIA attributes for screen readers
  • Focus management
  • Tailwind CSS styling

Database Expert#

Request:

Use bootspring_agent with agent="database-expert" and task="Design a schema for a multi-tenant SaaS with row-level security"

Response includes:

  • Prisma schema with tenant isolation
  • Row-level security patterns
  • Index recommendations
  • Query patterns for tenant filtering

Security Expert (Review Mode)#

Request:

Use bootspring_agent with agent="security-expert", task="Review this login endpoint", and outputFormat="review"

Response includes:

  • Security findings categorized by severity
  • Specific code fixes
  • OWASP compliance check
  • Recommendations for improvement

Agent Selection Guide#

ScenarioRecommended Agent
Building React/Next.js componentsfrontend-expert
Creating API endpointsbackend-expert, api-expert
Designing database schemasdatabase-expert
Setting up CI/CD pipelinesdevops-expert
Security review or authsecurity-expert
Writing teststesting-expert
Performance issuesperformance-expert
Mobile developmentmobile-expert
Design systems, accessibilityui-ux-expert
High-level architecturearchitecture-expert
Code quality reviewcode-review-expert
Vercel deploymentvercel-expert
Stripe/paymentspayment-expert
Email systemsemail-expert
AI/ML integrationai-integration-expert

Agent Chaining#

Chain multiple agents for complex tasks:

1. Use bootspring_agent with agent="architecture-expert" task="Plan the authentication system" 2. Use bootspring_agent with agent="database-expert" task="Design the user and session schema" 3. Use bootspring_agent with agent="backend-expert" task="Implement the auth endpoints" 4. Use bootspring_agent with agent="frontend-expert" task="Build the login and signup forms" 5. Use bootspring_agent with agent="security-expert" task="Review the implementation" 6. Use bootspring_agent with agent="testing-expert" task="Write tests for the auth system"

Configuration#

Custom Agent Instructions#

Loading code block...

Enable/Disable Agents#

Loading code block...

Error Handling#

ErrorCauseSolution
AGENT_NOT_FOUNDInvalid agent nameCheck bootspring agent list
AGENT_TIER_REQUIREDAgent requires higher tierUpgrade to Pro or Team
TASK_REQUIREDNo task providedInclude task parameter
CONTEXT_TOO_LARGEContext exceeds limitReduce context size

Best Practices#

1. Be Specific#

Instead of:

Use frontend-expert to help with the form

Try:

Use frontend-expert to create a multi-step checkout form with Stripe Elements, address validation, and proper error handling

2. Provide Context#

Include relevant details:

Use database-expert with context="We have 1M users and need to query by email frequently" to design the user schema

3. Use the Right Output Format#

  • guidance - Explanations and recommendations
  • code - Primarily code with comments
  • review - Security/code review format
  • plan - Step-by-step implementation plan

4. Chain for Complex Tasks#

Don't try to do everything with one agent. Break complex tasks into steps using multiple specialized agents.