Reset after failures, blockers, or when stuck. Use when execution hits a wall, something breaks, or we need to step back and reassess. This is the fifth system in the 5-system framework.

**Core Principle:** Commit to excellence. Work at 150% — 100% core delivery + 50% strengthening through verification, alternatives, and risk awareness.

[52] EXECUTE. Three-stage refactoring workflow: (1) iterative research of refactor/modularization options, (2) plan + risk/edge-case analysis + Scope150 validation, then implement with tests after user confirmation, and (3) apply Scout105 cleanup protocol. Use when asked to refactor, modularize, or restructure code safely.

Create your traffic engineering skill in one prompt, then learn to improve it throughout the chapter

Complete your dapr-deployment skill with actor and workflow patterns, validate code generation, and document your Digital FTE component

Complete browser automation with Playwright. Auto-detects dev servers, writes clean test scripts to /tmp. Test pages, fill forms, take screenshots, check responsive design, validate UX, test login flows, check links, automate any browser task. Use when user wants to test websites, automate browser interactions, validate web functionality, or perform any browser-based testing.

5W1H Decision Framework Tool. Use for: (1) Systematic decision-making before creating todos, (2) Preventing duplicate implementation, (3) Detecting avoidance behavior, (4) Ensuring agile refactor compliance with executor/dispatcher separation

Perform iterative root cause analysis by asking "why" repeatedly until you reach the underlying cause.

Test features before users find bugs. Use when feature is built, before deploying, or when bugs reported. Covers manual testing, edge cases, cross-browser testing, and testing checklists for non-technical founders.

This skill generates normalized BigWig (.bw) tracks (and/or fold-change tracks) from BAM files for ATAC-seq and ChIP-seq visualization. It handles normalization (RPM or fold-change) and Tn5 offset correction automatically. What's more, this skill can help user visualize the signal profiles around TSS or target regions. Use this skill when you have filtered and generated the clean BAM file (e.g. `*.filtered.bam`).

[60] VALIDATE. Ensure new (staged and unstaged) changes are covered by tests at >70% and the full test suite is green. Use when asked to validate coverage for recent changes, add tests for modified code, or verify nothing else broke.

You've spent this chapter building mental models: the LLMOps lifecycle, training taxonomy, economic analysis, use case specification. All valuable knowledge—but knowledge fades. Six months from now, y

[61] VALIDATE. Comprehensive code quality check combining ESLint, TypeScript compilation, and unused code detection. Runs full lint suite with detailed error reporting and fix suggestions. Use before commits, after major changes, or when ensuring code quality standards.

[62] VALIDATE. Final self-check before delivery. Verify goal alignment, completeness, correctness, and identify residual risks. Produces quality score (0-100) and delivery status. Use when completing any significant work, before handoff, or when you need confidence that work is ready.

Create your llmops-fine-tuner skill from Unsloth documentation before learning fine-tuning theory

Create your model-alignment skill from TRL documentation before learning DPO theory

Create a reusable skill for evaluating fine-tuned models, benchmarking performance, and detecting quality regressions

Fix bugs systematically instead of guessing. Use when features break, users report errors, or tests fail. Covers reproducing bugs, gathering diagnostic info, and working with AI tools to fix issues efficiently for non-technical founders.

[70] CLOSE. Manage task lifecycle with correct statuses and no duplicates. Ensure each task has one status, artifacts match claims, and no task appears in multiple places. Use when managing task lists, updating progress, or ensuring task tracking integrity across systems.

Create your agent-integration skill from OpenAI SDK and LiteLLM documentation before learning framework integration

[71] CLOSE. Quick, safe cleanup after completing a milestone. Fix objective issues only (syntax errors, dead code, poor naming). Must be <5% of main task time, <30 seconds per fix, and reversible. Use after key points, not after every small change.

Consolidate all Part 8 skills into a production-ready llmops-fine-tuner skill

**Core Principle:** Never lose progress. Save context before ending, restore context when starting.

[73] CLOSE. Record and maintain Lessons in MEMORY.md after a problem is solved or the user confirms success. Use when capturing a new lesson, moving lessons through the pipeline, or enhancing Project Architecture Quick Reference with new insights.

[74] CLOSE. Quick checkpoint during active work when context is running low. Use multiple times per development cycle to preserve progress and lessons. Lighter than close-session — no full handoff needed. Triggers on 'save progress', 'checkpoint', 'context low', or automatically when nearing token limits.

Create pixelated borders, shadows, and effects for 8-bit retro components. Apply when building 8-bit styled UI components that need authentic pixel art aesthetics.

The differential-region-analysis pipeline identifies genomic regions exhibiting significant differences in signal intensity between experimental conditions using a count-based framework and DESeq2. It supports detection of both differentially accessible regions (DARs) from open-chromatin assays (e.g., ATAC-seq, DNase-seq) and differential transcription factor (TF) binding regions from TF-centric assays (e.g., ChIP-seq, CUT&RUN, CUT&Tag). The pipeline can start from aligned BAM files or a precomputed count matrix and is suitable whenever genomic signal can be summarized as read counts per region.

Track production app health and catch issues before users complain. Use after deploying, to check app status, or when investigating user reports. Covers error tracking, uptime monitoring, and metrics for non-technical founders.

Generate web assets including favicons, app icons (PWA), and social media meta images (Open Graph) for Facebook, Twitter, WhatsApp, and LinkedIn. Use when users need icons, favicons, social sharing images, or Open Graph images from logos or text slogans. Handles image resizing, text-to-image generation, and provides proper HTML meta tags.

9D product development framework

Annual security report aggregation and analysis. USE WHEN annual reports, security reports, threat reports, industry reports, update reports, analyze reports, vendor reports, threat landscape.

name: "AppleFoundationModels"

Protocol fuzzer script development using boofuzz framework. Use when creating network protocol fuzzers, defining protocol PDUs (Protocol Data Units), writing mutation-based fuzzing scripts, or contributing to the boofuzz project. Covers binary protocol definitions, session management, crash detection, and reporting findings.

---name: cell_agent

Generate AIAG-format Control Plans linked to PFMEAs with proper inspection methods, frequencies, and reaction plans. Supports Prototype/Pre-launch/Production phases. USE WHEN user says 'control plan', 'CP', 'inspection plan', 'process control', 'sampling frequency', 'reaction plan', or 'special characteristics'. Integrates with PFMEA, AutomotiveManufacturing, and MSA skills.

Structured 8D problem solving for customer complaints and quality issues. D0-D8 phases with containment, root cause analysis, and escape point identification. USE WHEN user says '8D', 'eight disciplines', 'customer complaint', 'corrective action', 'root cause analysis', 'containment', 'escape point', or 'problem solving report'.

Agent evaluation framework based on Anthropic's best practices. USE WHEN eval, evaluate, test agent, benchmark, verify behavior, regression test, capability test. Includes three grader types (code-based, model-based, human), transcript capture, pass@k/pass^k metrics, and ALGORITHM integration.

First principles analysis. USE WHEN first principles, fundamental, root cause, decompose. SkillSearch('firstprinciples') for docs.

Conduct Gage R&R studies and validate measurement systems per AIAG MSA manual. Covers variable and attribute studies, acceptance criteria, and calculation methods. USE WHEN user says 'MSA', 'gage R&R', 'GR&R', 'measurement system', 'repeatability', 'reproducibility', 'attribute agreement', or 'ndc'. Integrates with ControlPlan, SPC, and AutomotiveManufacturing skills.

9P2000 protocol implementation for .NET - connect to 9front/Plan9, serve files, CPU remote shell, authentication servers

Generate AIAG-VDA compliant Process FMEAs with proper Severity/Occurrence/Detection ratings, RPN calculations, and countermeasure recommendations. Covers MNMUK departments (Machine Shop, Damper, LVA, FML). USE WHEN user says 'PFMEA', 'FMEA', 'failure mode', 'risk analysis', 'RPN', 'severity occurrence detection', or 'process risk assessment'. Integrates with AutomotiveManufacturing, ControlPlan, and A3criticalthinking skills.

PheroPath allows you to attach context (DANGER, TODO, SAFE, INSIGHT) to files using Extended Attributes (or a sidecar fallback).

UTF-8 JSON file I/O utilities to avoid Windows encoding issues (CP-1252 vs UTF-8)

Security reconnaissance. USE WHEN recon, reconnaissance, bug bounty, attack surface. SkillSearch('recon') for docs.

Skill para decomposição segura de arquivos monolíticos. Análise estrutural profunda, mapeamento exaustivo de dependências, extração incremental com zero quebra de lógica. Exige entendimento completo do negócio antes de qualquer ação.

Security news aggregation from tldrsec, no.security, and other sources. USE WHEN security news, security updates, what's new in security, breaches, security research, sec updates. SkillSearch('secupdates') for docs.

Implement SPC charting, process capability analysis, and control chart interpretation. Covers control chart selection, capability indices, and out-of-control rules. USE WHEN user says 'SPC', 'Cpk', 'Ppk', 'control chart', 'process capability', 'X-bar R', 'statistical control', or 'out of control'. Integrates with ControlPlan, MSA, and AutomotiveManufacturing skills.

Comprehensive guide to UI/UX design principles, architectural patterns, and animation techniques for building high-quality iOS apps with Swift.

**Before executing, check for user customizations at:**

Web security assessment. USE WHEN web assessment, pentest, security testing, vulnerability scan. SkillSearch('webassessment') for docs.