claude-code-deprecated

Accessibility audit against WCAG 2.1 AA standards. Invoke for: "accessibility", "a11y", "WCAG", "screen reader", "aria labels", "color contrast", "keyboard navigation", "accessibility audit", "ADA compliance".

Write Architecture Decision Records (ADRs) to document significant technical decisions. Invoke for: "write ADR", "document this decision", "architecture decision record", "why did we choose X", "decision log", "record this choice", "ADR for".

Design and implement multi-agent orchestration systems. Invoke for: "multi-agent system", "orchestrate agents", "agent pipeline", "agent workflow", "how should agents communicate", "agent coordination", "task delegation to agents", "build an agent system".

AI safety review: alignment, misuse prevention, bias, fairness, and responsible AI. Invoke for: "AI safety review", "bias check", "fairness audit", "alignment", "responsible AI", "misuse prevention", "AI ethics", "safety guardrails", "could this AI system be misused".

LLM and AI agent security: prompt injection, jailbreaks, agent defense, guardrails. Invoke for: "prompt injection", "LLM security", "agent security", "jailbreak defense", "AI safety audit", "system prompt leakage", "adversarial inputs", "AI pipeline security", "tool call validation", "LLM guardrails", "model security", "is my prompt safe".

Improve algorithmic efficiency — find better algorithms and data structures. Invoke for: "algorithm improvement", "better algorithm", "O(n²) is too slow", "data structure choice", "optimize this loop", "algorithmic complexity", "time complexity".

Design clean, consistent REST or GraphQL APIs with OpenAPI specs. Invoke for: "design this API", "API spec", "OpenAPI", "REST endpoints", "API review", "API versioning", "endpoint design", "HTTP API", "what endpoints do I need".

Generate and maintain API documentation. Invoke for: "document this API", "API docs", "OpenAPI spec", "Swagger", "endpoint documentation", "API reference", "document these endpoints", "API documentation missing".

Application security: OWASP Top 10, secure code review, secure SDLC. Auto-invoke on any code review request. Trigger for: "code security review", "OWASP audit", "XSS check", "SQL injection", "secure code review", "SAST", "dependency vulnerabilities", "auth review", "input validation", "secure coding", "review this code" (security angle), "is this safe", "any security issues".

Create ASCII architecture and system diagrams. Invoke for: "draw architecture diagram", "system diagram", "architecture overview", "visualize the system", "component diagram", "data flow diagram", "sequence diagram", "C4 diagram".

System design and architecture planning. Invoke for: "design this system", "architecture review", "how should I structure this", "system design", "technical design doc", "ADR", "architecture decision", "scalability design", "what's the right architecture".

Optimize async/concurrent code for throughput and correctness. Invoke for: "async issue", "concurrent", "race condition", "await optimization", "run in parallel", "asyncio", "Promise.all", "thread safety", "deadlock", "async bottleneck", "too many awaits", "sequential when should be parallel".

Design and verify backup and disaster recovery strategies. Invoke for: "backup strategy", "disaster recovery", "backup verification", "restore test", "RTO", "RPO", "data loss prevention", "backup schedule", "backup plan".

Identify and resolve project blockers autonomously. Invoke for: "I'm blocked", "this is blocking me", "unblock this", "resolve blocker", "stuck on X", "how do I get past this", "blocker". Autonomously diagnoses and resolves blockers without waiting for user hand-holding.

Socratic requirements refinement before writing any code. Surfaces assumptions, edge cases, tradeoffs, and hidden complexity through structured questioning. Invoke for: "brainstorm", "think through this", "what am I missing", "requirements unclear", "explore the design space", "what are the edge cases", "help me think through", "refine requirements", "what questions should I ask first". Inspired by Superpowers (obra/superpowers) /brainstorm phase.

Proactively hunt for bugs, edge cases, and failure modes before they hit production. Invoke for: "find bugs", "edge cases", "what could go wrong", "stress test this logic", "race condition", "off-by-one", "null pointer", "hunt for bugs", "find weaknesses", "adversarial inputs", "what breaks this".

Analyze and reduce JavaScript/Python bundle sizes. Invoke for: "bundle size", "reduce bundle", "tree shaking", "code splitting", "large dependency", "bundle too big", "import optimization", "lazy loading".

Design and implement caching strategies for performance. Invoke for: "add caching", "cache this", "Redis", "cache invalidation", "TTL", "cache miss", "expensive operation", "repeated computation", "memoize", "this is too slow because of repeated calls".

Low-risk conservative mode: extra confirmation before destructive ops, minimal blast radius, no assumptions. Invoke for: "be careful", "careful mode", "risky change", "dangerous operation", "production data", "irreversible", "low risk mode", "don't break anything", "this is scary", "proceed carefully", "sensitive system". Inspired by gstack /careful and /guard.

Chain-of-draft structured prompting: iteratively refine a draft through multiple focused critique-and-improve cycles. Invoke for: "chain of draft", "iterative refinement", "improve this through drafts", "draft and critique", "structured refinement", "progressive draft improvement", "multi-pass writing", "draft → critique → improve". Uses minimal token efficient drafts (CoD pattern from Xu et al. 2025).

Maintain and update the project CHANGELOG.md keeping it current. Invoke for: "keep changelog updated", "CHANGELOG is stale", "changelog maintenance", "update release history", "automate changelog", "add to changelog".

Generate and maintain CHANGELOG.md from git history. Invoke for: "update changelog", "what changed", "release notes", "changelog entry", "commit history summary", "CHANGELOG", "what's new in this release", "summarize changes since last release".

Design and optimize CI/CD pipelines for automated testing, building, and deployment. Invoke for: "CI/CD pipeline", "GitHub Actions", "GitLab CI", "build pipeline", "automate deployment", "continuous integration", "automated testing pipeline", "deploy on push", "pipeline config".

Full security orchestration — runs the complete security team as white-team orchestrator. Invoke for: "run a security audit", "check everything security", "full security review", "CISO review", "security assessment", "what's our security posture", "scan the whole project". Coordinates all 16 security sub-skills and produces a consolidated severity-ranked report.

Cloud infrastructure security and architecture (AWS/GCP/OCI). Invoke for: "cloud security", "IAM policies", "S3 bucket permissions", "cloud config review", "infrastructure security", "cloud hardening", "security groups", "network ACL", "cloud misconfiguration", "AWS security review".

Thorough code review for quality, correctness, performance, and maintainability. Auto-invoke when: "review this code", "check my code", "PR review", "code quality", "any issues with this", "look at this implementation", "is this good", "feedback on code". Checks logic errors, edge cases, performance, readability, test coverage, naming.

Research and analyze competitors and alternatives. Invoke for: "competitive analysis", "compare to competitors", "what are the alternatives", "market analysis", "how does X compare to Y", "competitor research", "SWOT analysis".

Design concurrent and parallel systems correctly. Invoke for: "concurrency design", "thread safety", "parallel processing", "worker pool", "queue", "producer consumer", "concurrent writes", "data consistency with concurrent access".

Show what changed between sessions or git refs as a structured context summary. Invoke for: "context diff", "what changed since last session", "diff since main", "what's new since yesterday", "changes since checkpoint", "session diff", "summarize changes since", "what did I change", "context since last commit". Inspired by russbeye/claude-memory-bank /context-diff pattern.

Analyze and reduce cloud and infrastructure costs. Invoke for: "reduce cloud costs", "cost optimization", "AWS cost", "cloud bill too high", "right-sizing", "idle resources", "spot instances", "reserved instances", "LLM cost too high".

Creates new skills or agents from scratch. Invoke proactively when: a repeatable workflow is identified, user says "make a skill", "create an agent", "turn this into a skill", "automate this", "save this workflow", "I keep doing this". If a workflow has been repeated twice, suggest capturing it. Decision: if task needs autonomous multi-step execution → agent; if it's a knowledge/instruction pack → skill.

Set up cron jobs and scheduled tasks. Invoke for: "schedule this", "run every day", "cron job", "scheduled task", "automation schedule", "run weekly", "cron expression", "set up recurring task". Also integrates with Claude Code's /schedule skill.

Design and build data pipelines for ingestion, transformation, and storage. Invoke for: "data pipeline", "ETL", "data ingestion", "process this data", "batch processing", "stream processing", "data transformation", "data workflow".

Curate, clean, and prepare datasets for AI training and evaluation. Invoke for: "clean this dataset", "prepare training data", "dataset curation", "deduplicate data", "label this data", "data quality", "prepare eval set", "filter bad examples".

Design normalized, production-ready database schemas from requirements. Invoke for: "design database schema", "data model", "entity relationship", "ER diagram", "normalize tables", "design tables", "create schema", "database design review", "foreign keys", "relational model", "schema planning".

Optimize database queries, indexes, and schema for performance. Invoke for: "slow query", "add index", "query optimization", "N+1 problem", "database performance", "schema review", "explain plan", "connection pooling", "query too slow".

Database security, encryption, and access control. Invoke for: "database security", "SQL injection review", "DB permissions", "encryption at rest", "connection string security", "database hardening", "query audit", "backup encryption", "db access review", "stored procedure security".

Systematic debugging: read error, trace root cause, fix autonomously. Invoke for: "this is broken", "debug this", "why is this failing", "error", "exception", "traceback", "not working", "fix this bug", "something's wrong", "it crashes". Never asks for hand-holding — reads logs, traces stack, identifies root cause, fixes.

Log and track technical and product decisions with rationale. Invoke for: "log this decision", "record why we chose X", "decision log", "we decided to", "document this choice", "capture this decision". Proactively suggest logging any significant technical choice made during a conversation.

Audit dependencies for vulnerabilities, license issues, and outdated packages. Invoke for: "dependency audit", "npm audit", "pip audit", "outdated packages", "CVE check", "license compliance", "supply chain security", "vulnerable dependency", "update packages".