manastalukdar
- 334 skills
- 0 followers
- last month last updated
- ▌
- ▌ Context Engineering · manastalukdarAudit and curate the 5-layer agent context hierarchy to prevent quality degradation mid-session — rules files, specs, source files, error output, and conversation history.
- ▌ Continuous Learning · manastalukdarExtract repeated patterns from archived sessions and offer to convert them into reusable skills
- ▌ Deployment Rollback · manastalukdarSafe deployment rollback with health checks and database migration reversal
- ▌ Performance Profile · manastalukdarPerformance profiling and bottleneck detection for Node.js, Python, and browser apps
- ▌
- ▌ Shipping And Launch · manastalukdarUnified pre-launch checklist + staged rollout strategy (0%→5%→25%→100%) + rollback triggers — consolidates deploy validation, canary monitoring, and rollback planning in one workflow.
- ▌ Architecture Diagram · manastalukdarGenerate architecture diagrams in Mermaid, PlantUML, or diagrams.net formats
- ▌
- ▌ Project Health · manastalukdarInternal health check skill for AI DevStudio project. Checks skill count consistency across docs, detects stale sessions, and identifies cache files to clean. Not user-facing.
- ▌ Skill Validator · manastalukdarInternal skill that validates a new or modified AI DevStudio skill against project standards. Checks YAML frontmatter, token optimization section, line limit, and naming conventions. Used by the code-reviewer agent during PR review.
- ▌ Prompt Injection Detect · manastalukdarScan skill files, CLAUDE.md, MCP configs, and agent prompts for prompt injection patterns
- ▌ Spec Driven Development · manastalukdarGated 4-phase workflow (Specify → Plan → Tasks → Implement) where nothing advances without explicit human review — surfaces assumptions before any code is written.
- ▌ Doubt Driven Development · manastalukdarSpawn a fresh adversarial reviewer mid-implementation for any non-trivial decision — branching logic, cross-module boundaries, irreversible operations — to catch what confirmation bias misses.
- ▌ Deprecation And Migration · manastalukdarFull deprecation lifecycle for APIs, features, or internal code — Hyrum's Law analysis, deprecation notices, migration path design, and parallel-run execution strategy.
- ▌ Source Driven Development · manastalukdarFetch official docs before writing any framework-specific code — implement to documented patterns, cite the source, and flag any deviation from the official approach.
- ▌ Incremental Implementation · manastalukdarEnforce vertical-slice discipline — implement one thin slice (~100 lines max), test it, verify it works, commit it, then repeat. Never write more than can be tested in a single step.
- ▌ Browser Testing With Devtools · manastalukdarUse Chrome DevTools MCP for live DOM inspection, console error capture, network request analysis, and paint profiling — runtime verification beyond what Playwright scripting covers.
- ▌ Observability And Instrumentation · manastalukdarInstrument code with structured logging, metrics, and OpenTelemetry tracing — define the on-call questions first, then work backward to the telemetry that answers them.
- ▌
- ▌ Remove AI Marks · manastalukdarStrip multi-vendor AI provenance marks from files and text — invisible Unicode (Layer A), statistical text watermarks via rewrite (Layer B), and C2PA/EXIF/XMP/container metadata from PNG/JPEG/WebP/SVG/PDF/DOCX/ODT/HTML/MD. Requires the watermarks-remover local service (Docker). Use when the user asks to strip watermarks, remove C2PA/Content Credentials, clean AI metadata, remove invisible Unicode, or runs /remove-ai-marks.
- ▌ Remove AI Tells · manastalukdarDetect AI-generated writing patterns in documents — reports filler openers, AI closers, hedge phrases, cliché transitions, template phrasing, AI vocabulary, rhetorical question-answer pairs, and overused intensifiers with a per-document AI-written percentage, then asks before removing anything.
- ▌
- ▌
- ▌ Signal Detector · manastalukdarAlways-on ambient skill that captures decisions, patterns, entity mentions, and original thinking from the current conversation and writes them to agent memory. Run in parallel — never blocks the main response.
- ▌ Visual AI Tells · manastalukdarDetect AI-generated UI patterns in CSS and component files — reports side-tab accents, gradient text, glassmorphism, bounce easing, template scaffolding, and monoculture font choices, then asks before removing anything
- ▌ AI System Review · manastalukdarReview LLM application code for context assembly quality, retrieval integration, prompt construction, output validation, cost controls, and failure handling
- ▌
- ▌ Debug Root Cause · manastalukdarRoot cause analysis with dependency tracing and call stack analysis
- ▌
- ▌ First Principles · manastalukdarDecompose a problem to its irreducible truths before proposing solutions — strip inherited assumptions, rebuild from fundamentals, surface the real constraint
- ▌
- ▌
- ▌ Setup Pre Commit · manastalukdarInstall Husky pre-commit hooks with lint-staged (Prettier), type checking, and test automation. Auto-detects package manager. Use when adding pre-commit hooks, setting up Husky, configuring lint-staged, or adding commit-time formatting and type checking.
- ▌ Webpack Optimize · manastalukdarWebpack/Vite/esbuild configuration optimization for build performance
- ▌
- ▌ API Test Generate · manastalukdarAuto-generate comprehensive API tests for REST and GraphQL endpoints with request/response validation
- ▌
- ▌ Component Library · manastalukdarScaffold component library structure with Storybook and documentation
- ▌
- ▌ Cache Strategy · manastalukdarImplement caching strategies for HTTP, service workers, and memoization
- ▌
- ▌ Context Budget · manastalukdarReport estimated context window utilization and flag when approaching the compaction threshold
- ▌ Deepen Modules · manastalukdarSurface architectural deepening opportunities — find shallow modules and refactor them into deep ones using precise vocabulary (module, seam, adapter, depth, leverage, locality). Use when refactoring architecture, reducing coupling, or improving testability.
- ▌ Domain Context · manastalukdarCreate and maintain a CONTEXT.md domain glossary and ADRs for the project — establishes shared vocabulary between developers, domain experts, and AI. Use when starting a new project, refining domain language, or resolving terminology ambiguities.
- ▌ Git Guardrails · manastalukdarInstall a Claude Code PreToolUse hook that blocks dangerous git operations (force push, reset --hard, clean -f, branch -D, checkout ., restore .) before they execute. Use when setting up git safety hooks or preventing destructive git commands in Claude Code.
- ▌
- ▌ Infrastructure · manastalukdarInfrastructure as Code generation for Terraform, CloudFormation, and Pulumi
- ▌
- ▌ Predict Issues · manastalukdarProactive problem identification through code analysis and pattern detection
- ▌ Query Optimize · manastalukdarSQL/NoSQL query optimization with N+1 detection and index recommendations
- ▌ Session Resume · manastalukdarResume previous work from archived session with full context restoration
- ▌
- ▌
- ▌ Context Restore · manastalukdarRestore work-in-progress context saved by /context-save — finds the most recent WIP commit on the current branch and unpacks it back to the working tree so you can resume exactly where you left off. Inspired by gstack's context-restore skill.
- ▌ Parallel Agents · manastalukdarManage multiple Claude instances for complex tasks with coordination
- ▌ Postman Convert · manastalukdarConvert Postman collections to automated tests with environment preservation
- ▌ Project Onboard · manastalukdarRun a structured onboarding interview to generate USER.md and PROJECT.md — capturing developer preferences, project constraints, access policies, and working style so every future agent session starts with full context.
- ▌ Prompt Engineer · manastalukdarDesign, optimize, and evaluate prompts for LLM applications — system prompts, few-shot examples, chain-of-thought, structured outputs, and caching strategy
- ▌ UI Critique · manastalukdarDual-agent UX design review — Assessment A runs design-director heuristics while Assessment B runs visual pattern analysis; results synthesized into a prioritized critique snapshot
- ▌
- ▌ Cleanproject · manastalukdarRemove debug artifacts and temporary files safely with git checkpoint protection
- ▌ Context Save · manastalukdarSave current work-in-progress context as a git WIP commit — captures all staged and unstaged changes with a timestamped stash message so you can switch contexts safely and restore later. Inspired by gstack's context-save skill.
- ▌ Contributing · manastalukdarContribution readiness assessment analyzing project guidelines and requirements
- ▌ Decision Log · manastalukdarRecord architectural and design decisions as structured ADR files in docs/decisions/, with auto-indexing and link suggestions to related decisions
- ▌ Devex Review · manastalukdarAudit the developer experience of your project — setup friction, onboarding clarity, local dev loop speed, tooling consistency, and documentation gaps. Produces a DX scorecard and prioritized improvement list. Inspired by gstack's devex-review skill.
- ▌ Drawio Skill · manastalukdar bundleUse when the user requests diagrams, flowcharts, architecture diagrams, ER diagrams, UML / sequence / class diagrams, network topology, cloud architecture from Terraform or Kubernetes manifests, ML/DL model figures (Transformer/CNN/LSTM), mind maps, or any visualization. Also use proactively when explaining systems with 3+ components, complex data flows, or relationships that benefit from visual representation. Best suited when the diagram needs custom styling, rich shape vocabulary, swimlanes, or exportable images (PNG/SVG/PDF/JPG). Generates .drawio XML and exports locally via the native draw.io desktop CLI.
- ▌
- ▌ Execute Plan · manastalukdarExecute implementation plans in controlled batches with validation and checkpoints
- ▌
- ▌ Interview Me · manastalukdarExtract requirements one targeted question at a time before any plan or code starts — hypothesizes user intent, scores confidence, and stops at ~95% confidence.
- ▌ Issue Triage · manastalukdarDiscover, deduplicate, prioritize, and label incoming GitHub issues and discussions — produce a structured priority queue with one-line summaries and suggested labels. Designed to run with /loop 2h or on demand.
- ▌ Legacy Audit · manastalukdarMap modernization opportunities in a legacy codebase — identify unsupported dependencies, architecture seams, and migration paths using strangler fig, adapter, and parallel-run patterns
- ▌ Office Hours · manastalukdarRun a YC-style office hours session before writing code — six forcing questions that surface wrong premises, missing constraints, and better approaches so you build the right thing the first time.
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌ AI Governance · manastalukdarAudit an AI/ML system for governance controls — bias risk, explainability, model risk management, data lineage, fairness metrics, and regulatory alignment (EU AI Act, NIST AI RMF)
- ▌
- ▌ Debug Session · manastalukdarDocument debugging sessions with hypothesis tracking and knowledge base
- ▌
- ▌
- ▌ Pr Babysitter · manastalukdarHerd open pull requests through review, CI, and merge — surface aging PRs, trigger minimal-fix on CI failures, nudge for rebases, and escalate blocked ones. Designed to run with /loop 5m.
- ▌ RAG Implement · manastalukdarDesign and implement Retrieval-Augmented Generation systems — chunking strategy, embedding selection, vector store setup, retrieval pipeline, re-ranking, and evaluation
- ▌ Security Scan · manastalukdarComprehensive security analysis with vulnerability detection and remediation tracking
- ▌ Session Start · manastalukdarBegin documented development session using Claude Code CLI's memory system
- ▌ Sessions Init · manastalukdarOrganize and synchronize session directory structure based on project architecture
- ▌ Tdd Red Green · manastalukdarEnforce true RED/GREEN TDD workflow with fail-first testing methodology
- ▌
- ▌ Test Mutation · manastalukdarMutation testing to verify test quality with Stryker, mutmut, or go-mutesting
- ▌ Agent Evaluate · manastalukdarDefine behavioral contracts, run adversarial tests, and detect regressions for AI agents — invariants, edge cases, statistical analysis, and benchmark-production gap detection
- ▌ Assumption Map · manastalukdarMap and prioritize the riskiest assumptions behind a feature, product decision, or technical design — surface what would kill it if wrong, then propose the cheapest test for each
- ▌ UI Polish · manastalukdarFinal pre-ship UI quality pass — reads the persisted critique snapshot as a prioritized backlog and implements fixes from P0 down
- ▌ Chaos Test · manastalukdarAnalyze codebase for resilience gaps — missing retry logic, absent timeouts, circuit-breaker opportunities, single points of failure, and degradation-mode blind spots
- ▌
- ▌ Lighthouse · manastalukdarRun Lighthouse audits and implement performance, accessibility, and SEO fixes
- ▌ Post Merge · manastalukdarSweep for follow-up work after recent merges — TODOs added in merged PRs, stale feature flags, dead code, documentation gaps, and tech-debt tickets left by merge comments.
- ▌ Smoke Test · manastalukdarRun 8 post-session-start health checks to verify the project is in a working state — git clean, no broken imports, tests pass, key files present, and agent memory intact.