fs-dev

Use when starting a new app idea, exploring product concepts, or doing market research. Reads existing specs, conducts competitor analysis, and helps flesh out the app idea with differentiation strategies. Say "I have an app idea", "interview me about this product", or "help me differentiate from competitors".

Use when designing a feature, planning implementation, exploring approaches, or needing to think through a change before coding. Runs an interactive interview to understand requirements, assess complexity, explore alternatives, and create GitHub Issues. Say "let's brainstorm", "help me design", or "plan this feature".

Use when reviewing competitor analysis results, getting a quick market overview, or summarizing competitive landscape from previous app-interview sessions. Generates executive summary directly to terminal. Say "summarize competitors", "show competitive analysis", or "what did we learn about the market?".

Run a multi-agent consensus code review. Dispatches 5+ parallel agents and synthesizes findings. All agents have veto power unless --ignore-vetos is set.

Use when auditing codebase quality, finding security issues, identifying tech debt, checking test coverage, or scanning for improvements. Runs autonomous analysis across security, quality, performance, and documentation. Say "analyze this codebase", "find issues", "audit security", or "check code quality".

Use when setting up a project to follow fyrsmithlabs standards. Works for new or existing repos - detects state automatically. Validates against git-repo-standards, generates CLAUDE.md based on project type, and configures git-workflows. Say "init this repo", "initialize project", "set up standards", or "make this follow fyrsmithlabs standards".

Use when starting a new feature or task from scratch. Orchestrates the full workflow - assesses complexity, runs brainstorming interview, creates GitHub Issues, and offers worktree setup. The complete planning pipeline. Say "plan this feature", "start planning", or "I want to build X".

Run multi-agent research orchestration on a topic

Use when refining an existing specification document, drilling into technical details, exploring edge cases, or validating design decisions. Conducts deep-dive interview about implementation, UI/UX, concerns, and tradeoffs. Say "refine this spec", "drill into the details", or "let's validate this design".

Run a daily standup report for the current project. Queries GitHub for PRs/issues and contextd memory for yesterday's state. For automated org-wide standups, see the daily-standup.yml workflow.

Use when testing marketplace skills for compliance, validating skill behavior against scenarios, or running TDD-style pressure tests. Tests if agents follow skill guidance correctly and catches rationalizations. Say "test this skill", "run pressure tests", or "validate skill compliance".

Manage YAGNI/KISS enforcement - view status, adjust sensitivity, understand nudges, toggle on/off, manage whitelist, view debt dashboard. Say "show yagni status", "why did I get that nudge?", "turn off yagni", "whitelist a pattern".

Code quality analysis agent for multi-agent consensus review. No veto power. Analyzes logic errors, test coverage gaps, cyclomatic complexity, code duplication, pattern violations, and error handling gaps.

Reviews technical documentation for accuracy, completeness, clarity, and adherence to best practices. Use when reviewing research reports, README files, API docs, or any technical writing. Has veto power for critical documentation issues.

Go language specialist agent for multi-agent consensus review. Has VETO power on Go-specific issues. Analyzes gofmt compliance, naming conventions, error handling, control flow patterns, concurrency safety, interface design, and resource management using Effective Go principles.

Product owner agent for deep priority analysis, cross-project dependency mapping, and strategic recommendations. Use for complex prioritization decisions, historical pattern analysis, or platform-wide planning that goes beyond a simple standup.

Architecture research agent for design patterns, system structure, and code organization. Use when researching how to structure code, which patterns to apply, or analyzing existing architecture.

Competitive research agent for industry trends, competitor analysis, and market context. Use when researching how others solve similar problems, market expectations, or industry standards.

Orchestrates parallel research agents for comprehensive topic research. Dispatches technical, architectural, UX, security, and competitive agents based on query analysis, then synthesizes results into actionable reports.

Security research agent for security best practices, OWASP guidelines, and vulnerability patterns. Use when researching security implications, authentication/authorization patterns, or threat modeling.

Synthesis agent that consolidates findings from all research agents into a unified, actionable report. Use as the final step in research orchestration to produce the deliverable.

Technical research agent for APIs, libraries, frameworks, and implementation patterns. Use when researching technical implementation details, comparing technologies, or finding code examples.

UX research agent for user experience patterns, accessibility standards, and usability best practices. Use when researching UI/UX design decisions, accessibility compliance, or user interaction patterns.

Security analysis agent for multi-agent consensus review. Has VETO power on security issues. Analyzes injection vulnerabilities, authentication flaws, secrets exposure, supply chain risks, and OWASP Top 10 violations.

User experience analysis agent for multi-agent consensus review. No veto power. Analyzes breaking API changes, migration path clarity, error message quality, configuration complexity, and developer ergonomics.

Vulnerability analysis agent for multi-agent consensus review. Has VETO power on security issues. Analyzes CVEs in dependencies, outdated packages, dependency confusion risks, license compliance, and transitive dependency risks.

Agent artifact placement conventions. Use when creating plans, tasks, research, reviews, brainstorms, or any agent-generated files. Ensures artifacts go to docs/.claude/ (gitignored) instead of polluting the repo.

Use when assessing task complexity, determining workflow depth, asking "how complex is this?", "should I use a worktree?", "is this a big change?", or sizing work before planning. Analyzes 7 dimensions (scope, integration, infrastructure, knowledge, risk, testing, decomposability) to return SIMPLE, STANDARD, or COMPLEX tier with confidence scoring.

Multi-agent consensus code review with adaptive budgets, context-folding isolation, and progressive summarization. Use when reviewing PRs, directories, or code changes with 5-6 parallel reviewer agents.

Use when executing complex sub-tasks that need context isolation - creates branches with token budgets that auto-cleanup on return

Use when writing Go code, reviewing Go PRs, or asking "is this idiomatic Go?", "how should I structure this in Go?", "what's the Go way to do X?". Provides Effective Go principles for development guidance and code review reference. Covers Go 1.18+ generics, Go 1.22+ iterators, modern testing, and tooling.

Use when creating new repositories, reviewing existing repos for compliance, or enforcing repository naming, structure, documentation, and security standards. Applies to all fyrsmithlabs projects.

Use when working with git operations, PRs, or code reviews - enforces modern agentic git workflows with multi-agent consensus review using contextd for memory, context folding, and learning capture

Use when creating GitHub Issues, making an epic, creating sub-issues, setting up a project board, or converting plans into GitHub artifacts. Creates tier-appropriate structures - SIMPLE (single issue with checklist), STANDARD (epic + native sub-issues), COMPLEX (epic + native sub-issues + Projects v2 board with automation).

Use when setting up a project to follow fyrsmithlabs standards. Works for new or existing repos - detects state automatically. Validates against git-repo-standards, generates missing artifacts, bootstraps CLAUDE.md based on project type. Includes interactive configuration wizard for project setup.

Use when running daily standups, prioritizing work, tracking cross-project dependencies, or managing development flow. Synthesizes GitHub state with AI-powered analysis, DORA metrics, and WSJF prioritization. Say "/standup", "what should I work on?", or "show project status".

Multi-agent research orchestration for gathering technical, architectural, UX, security, and competitive insights. Use before brainstorm, during planning, or for troubleshooting.

Use when analyzing codebase for improvements, running discovery, finding issues, auditing security/quality/performance/compliance/accessibility/DX, identifying tech debt, or scanning for problems. Autonomous non-interactive analysis with lens filtering (security, quality, perf, docs, dx, compliance, a11y) and severity classification (CRITICAL, MAJOR, MINOR, SUGGESTION). Supports parallel lens execution, complexity scoring, trend analysis, and multi-format output (GitHub Issues, JIRA, Markdown).

YAGNI/KISS enforcement with structured nudges. Use when discussing code complexity, over-engineering concerns, or to understand a yagni nudge.