karim-bhalwani
- 53 skills
- 0 followers
- 7 hours ago last updated
- ▌ Ops · karim-bhalwani bundleCI/CD pipelines, GitHub Actions workflows, Docker configurations, deployment automation, and release management patterns. Use when building CI/CD, writing GitHub Actions, creating Docker configs, automating deployments, managing releases, or setting up infrastructure-as-code. DO NOT USE FOR: writing application-specific business logic (use implementer), designing high-level system architecture or topology (use architect), performing security audits (use guardian), or orchestrating data pipelines like Airflow DAGs (use data-engineering).
- ▌ LLM Mem · karim-bhalwani bundleBuild and maintain a project knowledge mem using LLMs. Triggers: ingesting sources into a mem, querying mem knowledge, linting mem health, 'add to mem', 'what do I know about', 'compile to mem', or any mention of 'project mem'. Use when persisting durable knowledge (patterns, decisions, post-mortems, research) beyond the current session. DO NOT USE FOR: project setup or context files (use context-engineer), session-scoped memory (use memory tool directly), documentation generation without mem structure, or LLM app design (use llm-app-patterns).
- ▌ Data Analyst · karim-bhalwani bundleAzure SQL Server and T-SQL query specialist covering natural language to SQL conversion, schema exploration, Data Vault 2.0 querying patterns, query optimization, and SSMS workflows. Use when writing SQL queries from natural language requests, exploring database schemas, navigating Data Vault warehouses (Hubs, Links, Satellites, PIT, Bridge, EffSats), optimizing T-SQL performance, or generating copy-ready SQL scripts for Azure SQL or SQL Server databases. DO NOT USE FOR: building ETL/ELT pipelines (use data-engineering), PySpark or dbt transformations (use data-engineering), deprecation audits (use data-deprecation-analysis), or non-SQL database work.
- ▌ Data Science · karim-bhalwani bundleComprehensive data science reference covering exploratory data analysis, statistical hypothesis testing, predictive modeling with scikit-learn/XGBoost, time series forecasting, and A/B experiment design. Use when profiling datasets, running hypothesis tests, building classification/regression models, forecasting time series, designing experiments, or interpreting model results. DO NOT USE FOR: ad-hoc SQL querying (use data-analyst), ETL/ELT pipeline construction (use data-engineering), LLM/RAG application design (use llm-app-patterns), MLOps and deployment automation (use ops), or pure code review (use guardian).
- ▌ Story Master · karim-bhalwani bundleHouses the scaffold_stories.py and verify_stories.py scripts used by the story-master agent (prompts/story-master.agent.md) to generate and validate STORIES.md backlog files. This skill is not directly invocable; it is a script-housing skill that the story-master agent shells out to. This skill provides only scaffolding and verification functionality; backlog decomposition is handled by the story-master agent, code review by guardian, and implementation planning by story-planner agent.
- ▌ Task Routing · karim-bhalwaniMulti-agent delegation protocol with sequentiality, decomposability, and cost-benefit checks derived from DeepMind/MIT research. Load when deciding whether to delegate a task to another agent or handle it yourself. Includes coordination anti-patterns that degrade agent system performance. DO NOT USE FOR: executing delegated tasks (use subagent-execution), creating implementation plans (use concise-planning), actual code work, or single-step tasks that don't need delegation.
- ▌ Story Planner · karim-bhalwani bundleHouses scaffold_plan.py, scaffold_validation.py, verify_plan.py, and verify_validation.py used by the story-planner agent (prompts/story-planner.agent.md) to generate and validate per-story planning artifacts (US-NN-PLAN.md and US-NN-VALIDATION.md). This skill is not directly invocable; it is a script-housing skill that the story-planner agent shells out to.
- ▌ Data Narrative · karim-bhalwaniTurns a raw dataset into a structured, evidence-grounded Markdown report where every claim traces back to the code or source that produced it. Orchestrates four specialist roles: Detective (external context), Analyst (exhaustive data profiling via data-science skill), Editor (narrative angle + report.md), and Inspector (provenance binding → inspector.json). USE FOR: any dataset-to-report workflow where auditability and claim traceability matter - analytical briefings, stakeholder reports, data-backed narratives. DO NOT USE FOR: SQL querying (use data-analyst), pipeline construction (use data-engineering), standalone EDA without a deliverable (use data-science), or architecture diagrams (use excalidraw-diagram).
- ▌ Genai Security · karim-bhalwani bundleSecurity auditing for GenAI/LLM applications. Primary focus: OWASP Top 10 for LLMs (2025). Use for LLM-powered applications, RAG pipelines, AI agents, prompt templates, and generative AI integrations. For agentic risks, prompt injection patterns, MITRE ATLAS mappings, threat modeling, or structured red teaming methodology, load specific reference documents - outputs are analysis and findings, not formal penetration test reports. NOT for: general code security (use guardian), building LLM apps (use llm-app-patterns), runtime rules (use security-boundaries), or infrastructure.
- ▌ Data Engineering · karim-bhalwani bundleComprehensive data engineering reference covering Medallion architecture, Data Vault 2.0, PySpark optimization, dbt transformation patterns, data quality frameworks, and SQL optimization. Use when building ETL/ELT pipelines, designing data schemas, optimizing Spark/SQL performance, implementing data governance, or architecting lakehouse solutions. DO NOT USE FOR: ad-hoc SQL querying (use data-analyst), deprecation analysis (use data-deprecation-analysis), LLM/RAG pipeline design (use llm-app-patterns), or CI/CD configuration (use ops).
- ▌ Excalidraw Diagram · karim-bhalwani bundleGenerate structured, argumentative Excalidraw diagrams from natural language descriptions. Use when creating architecture diagrams, system visualizations, data flow diagrams, pipeline visualizations, workflow charts, or any technical illustration that outputs .excalidraw JSON files. DO NOT USE FOR: Mermaid or markdown diagrams, text-based documentation (use brainstorming or architect), code generation, or non-visual output.
- ▌ Holdout Validation · karim-bhalwani bundleStructural separation of test authorship from code authorship. Produces behavioral acceptance scenarios that implementation agents cannot see, evaluated independently by Guardian. Use when designing acceptance criteria, validating implementations against intent, or enforcing holdout-set discipline. DO NOT USE FOR: unit test writing (use guardian), implementation (use implementer), verifying task completion (use verification-before-completion), or any other test strategies beyond holdout validation, including integration and performance testing.
- ▌ Subagent Execution · karim-bhalwaniUse when executing an approved implementation plan by dispatching work to subagents. Enforces context isolation, two-stage review per task, and a formal status protocol. Load before orchestrating multi-task plans across subagents. DO NOT USE FOR: deciding whether to delegate (use task-routing), creating the plan itself (use concise-planning), single-task execution you can handle directly, or code review (use guardian).
- ▌ Security Boundaries · karim-bhalwaniPrompt injection defense rules, trust boundaries, and agent-specific security notes. Load when reviewing untrusted content, building security-sensitive features, or auditing code for injection vulnerabilities. Extracted from global instructions to reduce auto-loaded context. DO NOT USE FOR: LLM application security auditing (use genai-security), general code review (use guardian), OWASP web vulnerabilities without prompt injection (use guardian), or system architecture.
- ▌ Systematic Debugging · karim-bhalwaniUse when a bug, error, or unexpected behavior needs investigation. Enforces evidence-first root cause analysis before any fix attempt. Load before any debugging session - covers software-related failure modes related to active errors and bugs (data pipelines, API errors, LLM misbehavior). DO NOT USE FOR: code review without a specific error (use guardian), implementing known fixes (use implementer), performance profiling without failures (use guardian), or architecture design (use architect).
- ▌ Editor · karim-bhalwaniEditor sub-role of the data-narrative skill. Reads analyst.json and detective.json, selects a single compelling narrative angle, ranks findings by story priority, and drafts the structured report.md. Loaded and invoked by the data-narrative orchestrator only.
- ▌ Analyst · karim-bhalwaniAnalyst sub-role of the data-narrative skill. Delegates exhaustive dataset profiling to the data-science skill, then emits analyst.json - a structured list of findings where every finding carries a pointer to the script and line that produced it. Loaded and invoked by the data-narrative orchestrator only.
- ▌ Detective · karim-bhalwaniDetective sub-role of the data-narrative skill. Gathers external context that the dataset alone cannot supply - domain background, relevant events, definitions, and reference sources - using web search. Emits detective.json with every item tagged to its source URL. Loaded and invoked by the data-narrative orchestrator only.
- ▌ Inspector · karim-bhalwani bundleInspector sub-role of the data-narrative skill. Links every claim in report.md back to its upstream evidence (Analyst code line or Detective source URL). Produces inspector.json - the machine-verifiable provenance manifest for the narrative. Loaded and invoked by the data-narrative orchestrator only.
- ▌ Thinker · karim-bhalwani bundleSpecialized in structured reasoning, mental scaffolding, and breaking down complex problems. Use at the start of every task to ensure deep, transparent, and auditable thinking before taking action.
- ▌ Guardian · karim-bhalwani bundleSpecialized in quality assurance, security auditing, and performance optimization. Use when conducting code reviews, hunting bugs, scanning for vulnerabilities, profiling performance, ensuring security standards, or validating implementation quality.
- ▌ Architect · karim-bhalwani bundleSpecialized in system architecture, API contracts, and modular black-box design. Use when creating technical specifications, designing system primitives, defining module boundaries and contracts, reviewing architecture for replaceability, or mapping system dependencies.
- ▌ Navigator · karim-bhalwani bundleSpecialized in codebase exploration, finding file paths, tracing execution logic, and mapping dependencies. Use when exploring unfamiliar codebases, finding where code is located, tracing function calls, understanding module structure, or mapping system dependencies.
- ▌ QA Tester · karim-bhalwani bundleExpert in automated QA testing, E2E test automation, visual regression testing, and intelligent test generation. Use when building automated test suites, validating user workflows, testing external integrations, setting up test infrastructure, or implementing AI-powered test discovery.
- ▌ Researcher · karim-bhalwani bundleSpecialized in verifying technical facts, fetching external documentation, and validating syntax to prevent hallucinations. Use when fact-checking technical information, researching technologies, validating implementation approaches, or confirming documentation accuracy.
- ▌ Implementer · karim-bhalwani bundleSpecialized in high-quality implementation, test-driven development, and clean code practices. Use when writing features, fixing bugs, implementing architectural specifications, refactoring code, or ensuring code quality.
- ▌ Ops Manager · karim-bhalwani bundleSpecialized in CI/CD, Infrastructure-as-Code, release planning, and technical documentation. Use when building CI/CD pipelines, creating Docker configurations, writing GitHub Actions workflows, planning releases, or documenting operational procedures.
- ▌ Proj Memory · karim-bhalwani bundleSet up and maintain structured project memory systems that track bugs with solutions, architectural decisions, key project facts, and work history. Use when setting up new projects, documenting decisions, tracking bugs and solutions, or maintaining project knowledge across sessions.
- ▌ Data Modeler · karim-bhalwani bundleDesign and implement data pipelines using Medallion architecture, Data Vault 2.0, and Spark optimization. Use when building ETL/ELT processes, designing data schemas, optimizing Spark performance, implementing data governance, or architecting lakehouse solutions.
- ▌ Orchestrator · karim-bhalwani bundleSpecialized in managing project-wide constraints, context tiers, and coordination between specialized skills. Use when interpreting project rules, coordinating complex workflows, managing context and dependencies, or overseeing multi-skill project execution.
- ▌ Brainstorming · karim-bhalwani bundleExplore user intent, requirements, and design before implementation. Use when turning ideas into designs, validating requirements through dialogue, exploring multiple approaches, or creating design documentation for features, components, and functionality changes.
- ▌ State Manager · karim-bhalwani bundleSpecialized in maintaining project logs, tracking decisions, and validating implementation against original specifications. Use when documenting decisions, maintaining project history, validating work completeness, auditing implementation, or tracking project state.
- ▌ Prompt Library · karim-bhalwani bundleCurated collection of high-quality prompts for various tasks—role-based templates, analysis frameworks, code review prompts, and brainstorming structures. Use when building prompts for specific roles, needing analysis or brainstorming frameworks, creating code review prompts, or searching for proven prompt patterns and structures.
- ▌ Concise Planning · karim-bhalwani bundleGenerate clear, actionable, and atomic checklists for coding tasks with minimal interaction and reasonable assumptions. Use when creating implementation plans, breaking down tasks into steps, generating task checklists, or planning project workflows.
- ▌ Context Engineer · karim-bhalwani bundleInitialize project infrastructure, generate context files, and maintain the Single Source of Truth. Use when setting up new projects, auditing project environments, generating context documentation, maintaining project memory, or ensuring consistency across the codebase.
- ▌ JSON To Pydantic · karim-bhalwani bundleConvert JSON data snippets into Python Pydantic data models with proper type annotations and validation. Use when converting JSON schemas to Pydantic models, generating data validation models from JSON examples, or creating Python data classes.
- ▌ LLM App Patterns · karim-bhalwani bundleProduction LLM application patterns, architectures, and best practices. Covers RAG pipelines, agent architectures, prompt engineering, LLMOps, and production deployment patterns.
- ▌ Workflow Manager · karim-bhalwani bundleSpecialized in executing standardized project workflows and managing the lifecycle of planning artifacts. Use when managing project workflows, coordinating phases, ensuring SDLC consistency, or automating procedural tasks.
- ▌ Spark Optimization · karim-bhalwaniSpecialist in Apache Spark performance optimization—partitioning strategies, memory tuning, shuffle reduction, and job profiling for production systems. Use when optimizing Spark jobs, tuning performance, reducing costs, profiling applications, or scaling Spark workloads.
- ▌ Senior Data Engineer · karim-bhalwaniExpert in complex pipeline scaling, performance optimization, and architectural problem-solving. Use when solving high-performance systems challenges, optimizing data platforms, mentoring engineers, reviewing complex data architectures, or handling large-scale data engineering projects.
- ▌ Software Architecture · karim-bhalwani bundleGuide for quality focused software architecture. Use when designing software systems, reviewing architecture, analyzing code structure, planning refactors, or mentoring on architectural patterns.
- ▌ Web Design Guidelines · karim-bhalwani bundleReview UI code for Web Interface Guidelines compliance. Use when reviewing UI code, checking accessibility standards, auditing design consistency, or validating sites against best practices.
- ▌ Data Pipeline Engineer · karim-bhalwani bundleSpecialist in hands-on ETL/ELT design, implementation, and optimization using Airflow, Spark, Kafka, dbt, and data quality frameworks. Use when building data pipelines, orchestrating data workflows, optimizing data processing, implementing streaming solutions, or ensuring data quality throughout pipelines.
- ▌ Data Quality Frameworks · karim-bhalwani bundleSpecialist in data quality validation frameworks—Great Expectations, dbt tests, data contracts. Builds comprehensive data quality gates into pipelines for reliability and trust.
- ▌ Principal Data Engineer · karim-bhalwani bundleStrategic technical leadership in data platform design, architecture decisions, and engineering excellence. Use when making strategic architecture decisions, designing data platforms at scale, mentoring teams, or solving complex data engineering challenges.
- ▌ Product Manager Toolkit · karim-bhalwani bundleComprehensive toolkit for product managers including RICE prioritization, customer discovery, PRD templates, metrics frameworks, and go-to-market strategies. Use when prioritizing features, conducting customer research, defining requirements, or tracking product success.
- ▌ Top Web Vulnerabilities · karim-bhalwani bundleComprehensive reference for the OWASP Top 100 web vulnerabilities. Identifies vulnerable patterns, explains impacts, and provides remediation guidance across injection attacks, authentication flaws, data exposure, and advanced attack vectors.
- ▌ Data Deprecation Analysis · karim-bhalwani bundleSpecialist in analyzing access logs to identify unused data structures, technical debt, and legacy patterns. Use when auditing data warehouses for deprecation candidates, detecting stale tables/views, identifying legacy application dependencies, or planning safe data asset retirement.
- ▌ SQL Optimization Patterns · karim-bhalwani bundleSpecialist in SQL query optimization—index strategies, EXPLAIN analysis, query tuning. Transforms slow queries into fast ones through systematic diagnosis and targeted optimization.
- ▌ Github Workflow Automation · karim-bhalwani bundleAutomate GitHub workflows with AI-powered assistance for PR reviews, issue triage, CI/CD pipelines, and Git operations. Use when setting up automated code reviews, implementing issue labeling, creating smart CI/CD workflows, building deployment validation, automating Git operations, or configuring on-demand AI assistance bots.
- ▌ Dbt Transformation Patterns · karim-bhalwani bundleExpert in dbt (data build tool) for SQL-first analytics engineering—model organization, incremental strategies, testing, documentation, and lineage management. Use when building dbt projects, designing transformation models, implementing incremental strategies, testing data quality, or managing dbt documentation.
- ▌ Subagent Driven Development · karim-bhalwani bundleExecute implementation plans with independent tasks in the current session using fresh subagents with two-stage review after each task. Use when breaking down complex work into parallel tasks, managing subagent execution, or coordinating multi-step implementation.
- ▌ Verification Before Completion · karim-bhalwani bundleUse when about to claim work is complete, fixed, or passing—requires running verification commands and confirming output before making any success claims.