claude-code-project-best-practices

Monitors analysis-doc lifecycle in all three lanes — EMERGING→PRODUCTION promotion, PRODUCTION→RETIRING/RETIRED retirement (replacement-readiness as robust community/vendor elements mature), and follow-lane upkeep (followed canons' advance triggers + liveness, per ABSORPTION-MAP.md)

Validates new analysis documents against project quality standards. Trigger when user adds an analysis doc or says "review analysis".

Ensures SOURCES.md stays synchronized when adding new analysis documents or sources. Trigger when user adds a new analysis doc, says "update sources", "add source", or after pattern-reviewer identifies missing source entries. Enforces evidence tier classification and proper citation format.

Analyze recent git commits in a target project to surface patterns, identify improvement opportunities, and generate recommendations weighted by source authority and recency. Trigger when user says "analyze commits", "review my recent work", "what patterns do you see", or when evaluating a project for best-practice alignment.

Updates analysis documents with new version requirements and dependencies when Claude Code releases new features

Apply comprehensive quality review when user prepares content for publication (blog posts, articles, documentation, presentations). Trigger when user mentions "publish", "post", "article", "review this", "draft complete", "ready to share". Enforce evidence-based claims, intellectual honesty, professional voice, and balanced perspective.

Apply detection engineering quality standards when user writes SIEM rules, detection logic, or security monitoring queries. Trigger when user mentions "detection rule", "SIEM", "Sigma", "alert", "Splunk query", "KQL", or shares security detection code. Evaluate for accuracy, performance, evasion resistance, and operational quality.

Apply git best practices for commits, branches, worktrees, and collaboration when user performs version control operations. Trigger when user mentions "git", "commit", "branch", "merge", "push", "pull request", or prepares to commit code. Ensure clear commit messages, proper branching, and safe git operations.

Apply systematic hypothesis validation when user formulates research claims or proposes testable hypotheses. Trigger on "I hypothesize that...", "Can you validate this claim?", or research findings requiring evidence assessment. Use confidence scoring and evidence tier classification.

Apply Self-Evolution Algorithm for complex research and analysis. Trigger when user needs comprehensive analysis, high-stakes decisions, or requests "deep research", "multiple perspectives", "leave no stone unturned". Spawns parallel candidates (conservative/balanced/creative), refines each through judge feedback, and synthesizes best elements through crossover.

Systematically extract and document insights from research sources (papers, articles, documentation, talks). Trigger when user analyzes academic papers, reviews technical content, reads industry articles, or mentions "this paper", "according to research", "the article discusses". Capture concepts, evidence, hypotheses, and bibliography entries with proper attribution.

Apply 4-phase root cause debugging methodology when user encounters errors, bugs, test failures, or unexpected behavior in code. Trigger when user mentions "bug", "error", "failing", "not working", "debug", or shares error messages. Use systematic approach rather than guess-and-check.

Enforce Test-Driven Development (TDD) RED-GREEN-REFACTOR cycle when user writes code, implements features, or fixes bugs. Trigger when user mentions "implement", "add feature", "create function", "build", or starts coding without tests. Apply to Python, TypeScript, JavaScript, and all languages with testing frameworks.

Apply systematic threat modeling review when user designs systems, evaluates security architectures, or analyzes attack surfaces. Trigger when user mentions "threat model", "security review", "attack surface", "STRIDE", "trust boundary", or asks about potential threats. Use structured methodology to identify threats, assess risks, and recommend mitigations.

Apply comprehensive FRAME-ANALYZE-SYNTHESIZE methodology for deep analysis of complex problems, strategic decisions, architectural choices, or research questions. Trigger when user says "ultrathink", "deep analysis", "systematic analysis", "comprehensive evaluation", or asks to analyze complex multi-dimensional problems.

Regenerates INDEX.md to reflect current file structure. Trigger when user says "update index", "regenerate index", after adding/removing files, or when INDEX.md shows stale counts. Provides more control than the PostToolUse hook which runs automatically.

Surfaces hidden assumptions before complex tasks using Johari Window methodology. Trigger: "clarify", "what am I missing", or tasks involving 3+ files.