NXTG Forge

Update stale documentation based on code changes

Add a new feature with full agent orchestration

Display complete project state (zero-context-friendly)

Execute project tests with detailed analysis and reporting

Update Forge to latest — syncs past Claude Code bug #29071

Use this agent when metrics tracking, data analysis, or reporting is needed. This includes: implementing analytics events, building dashboards, analyzing usage patterns, creating reports, tracking KPIs, and measuring feature adoption. <example> Context: User wants to understand usage patterns. user: "Which features are used most?" assistant: "I'll use the analytics agent to analyze usage data and create a report." <commentary> Usage analysis and reporting is a analytics task. </commentary> </example> <example> Context: User needs to track a new metric. user: "Track how long each agent takes to complete tasks" assistant: "I'll use the analytics agent to implement timing instrumentation for agent tasks." <commentary> Metrics instrumentation is a analytics specialty. </commentary> </example>

Use this agent when API design, endpoint creation, or API integration work is needed. This includes: designing REST endpoints, implementing request/response handlers, adding validation, creating API middleware, integrating with external APIs, and generating OpenAPI specs. <example> Context: User needs a new API endpoint. user: "Add a PATCH endpoint for updating agent configuration" assistant: "I'll use the api agent to design and implement the endpoint with validation." <commentary> API endpoint creation with proper validation is a api task. </commentary> </example> <example> Context: User wants to integrate an external service. user: "Connect to the GitHub API to fetch repo stats" assistant: "I'll use the api agent to implement the GitHub API integration." <commentary> External API integration is a api specialty. </commentary> </example>

Use this agent when you need to implement features, write production-ready code, refactor existing code, or generate comprehensive tests. This agent should be invoked after a plan has been approved or when specific code implementation is required. <example> Context: The Forge Orchestrator has approved an implementation plan for a new authentication system. user: "The authentication plan looks good, let's implement it." assistant: "I'll use the Task tool to launch the builder agent to implement the authentication system according to the approved plan." <commentary> Since the user has approved an implementation plan, use the builder agent to transform the plan into production-ready code with tests and documentation. </commentary> </example> <example> Context: User needs to refactor a service class that has grown too large. user: "The UserService class is 500 lines and needs to be refactored into smaller pieces." assistant: "I'll use the Task tool to launch the builder agent to analyze and refactor the UserService class following SOLID principles." <commentary> Since refactoring work is needed, use the builder agent which specializes in code quality improvements and applying design patterns. </commentary> </example> <example> Context: User requests implementation of a specific feature with tests. user: "Implement a password reset flow with email verification." assistant: "I'll use the Task tool to launch the builder agent to implement the password reset flow with comprehensive tests and documentation." <commentary> Since specific code implementation is requested, use the builder agent to generate clean, maintainable, well-documented code with tests. </commentary> </example> <example> Context: Code review identified quality issues that need fixing. user: "The repository layer needs better error handling and type hints." assistant: "I'll use the Task tool to launch the builder agent to improve the repository layer with Result types and proper type annotations." <commentary> Since code quality improvements are needed, use the builder agent which applies best practices like Result types and strict type safety. </commentary> </example>

CEO Digital Twin for autonomous strategic decision-making. Use this agent when you encounter decisions about product direction, architecture, resource allocation, risk assessment, or when you need final approval on significant changes. This agent embodies the founder's vision and makes decisions with "ship fast, iterate faster" energy. ONLY escalate to human for CRITICAL decisions (Impact: CRITICAL + Risk: CRITICAL). For everything else, CEO-LOOP decides autonomously. Examples: - "Should we implement feature X or feature Y first?" → CEO-LOOP decides - "Is this architecture approach aligned with our vision?" → CEO-LOOP decides - "Should we ship this with known minor bugs?" → CEO-LOOP decides - "Strategic pivot: completely change product direction?" → Escalate to human

Use this agent when regulatory compliance, license auditing, or policy enforcement is needed. This includes: license compatibility checks, GDPR/privacy compliance, accessibility compliance (WCAG), code of conduct enforcement, and export control checks. <example> Context: User wants to check license compatibility. user: "Are all our dependencies license-compatible?" assistant: "I'll use the compliance agent to audit dependency licenses." <commentary> License auditing is a compliance task. </commentary> </example> <example> Context: User needs GDPR compliance check. user: "Does our data handling comply with GDPR?" assistant: "I'll use the compliance agent to review data handling for GDPR compliance." <commentary> Privacy compliance review is a compliance specialty. </commentary> </example>

Documentation standards and code-to-docs sync conventions for Forge projects — JSDoc/TSDoc annotations, README/CHANGELOG structure, auto-generated API reference, doc-tree layout, and staleness detection. Use when writing or reviewing docs, adding JSDoc to exported functions, structuring a docs/ tree, deciding what to document vs auto-generate, or running the Forge docs commands (/forge:docs-status, /forge:docs-audit, /forge:docs-update).

Backend implementation playbook — API endpoints, database models/migrations, auth, query optimization, and the framework-specific traps that break servers in production. Use when implementing or reviewing server-side code: FastAPI / Django / Flask / Express / NestJS / Go / Axum handlers, SQLAlchemy / Prisma / Mongoose / Tortoise models, JWT / OAuth / session auth, bcrypt / argon2 password hashing, rate limiting, N+1 query fixes, async/event-loop bugs, or when a handoff spec asks for "the backend" of a feature.

Knowledge for designing and building command-line interfaces — argument/flag parsing, subcommand structure, interactive prompts, colored and tabular output, progress feedback, exit codes, shell completion, and CLI UX. Use when adding or redesigning a CLI command, choosing a CLI framework (clap, Click/Typer, Commander, Cobra), fixing an argument-parsing or flag-naming bug, adding an interactive wizard or progress bar, or improving CLI error messages and help text.

Author and configure Claude Code subagents — valid YAML frontmatter (name, description, model, color, tools, isolation, memory, skills) plus a system-prompt body. Use when creating a new agent .md, fixing an agent that won't load or won't auto-delegate, choosing a model/color, deciding which tools an agent gets, or reviewing agent frontmatter for invalid/ignored fields.

Resilient third-party integration patterns — API clients, webhooks, MCP server wiring, retries/circuit-breakers, and idempotency. Use when connecting to an external service (Stripe/GitHub/Slack/S3/etc.), building or verifying a webhook receiver, adding retry/backoff or circuit-breaker resilience, registering an MCP server, or reviewing an integration for signature-verification and rate-limit gaps.

23 hooks across 4 events