Testing & Debugging for Claude Code

Coordinates accessibility, performance, API design, and dependency audit agents in parallel to produce a scored quality report. Use when running a pre-release gate or investigating multi-domain quality issues. Trigger with \"run a quality audit\", \"check this app for quality issues\".

Profiles applications, identifies bottlenecks (DB, I/O, memory, CPU), and produces optimization recommendations with benchmarked metrics. Use when an app is slow, traffic is scaling, or you need a pre-release performance baseline. Trigger with \"profile this service\", \"find performance bottlenecks\".

Implements and reviews React components using modern hooks, state management, and performance patterns (memoization, virtualization, code splitting) with TypeScript. Use when architecting components, debugging re-renders, or choosing a state strategy. Trigger with \"build this React component\", \"review my React code\".

Makes minimal, surgical edits to complex or large files with zero collateral damage — mapping dependencies, preserving invariants, and verifying syntax after each atomic change. Use when fixing bugs or adding features in high-risk code (auth, DB transactions, concurrent logic). Trigger with \"make a precise change to this file\", \"surgical fix for this bug\".

Manages Linux service lifecycle (start, stop, restart, health check, crash investigation) via systemd and the sm service manager, delegating all routing changes to geepers_caddy. Use when a service is down, crashing, or needs deployment. Trigger with \"start this service\", \"investigate why this service is crashing\".

Full-depth infrastructure health check for dr.eamer.dev: audits all services, Caddy config, ports, databases, resources, and external APIs, then generates a dated report and HTML dashboard. Use when something feels wrong or before major changes. Trigger with \"run system diagnostic\", \"full infrastructure audit\".

Validates all aspects of a project before deployment — config files, paths, permissions, ports, service status, and cross-service integration — then produces a categorized report with actionable fixes. Use when preparing to deploy or debugging mysterious failures. Trigger with "validate project", "pre-deploy check".

Runs hypothesis-driven threat hunts to surface attackers that evaded automated detection — IOC enrichment, compromise assessment, and hunting playbooks. Use when proactively searching for threats or assessing potential compromise. Trigger with \"run a threat hunt\", \"assess for compromise\".

Generates production-ready Kubernetes manifests with security contexts, resource limits, probes, HPA, and Ingress, plus kubectl troubleshooting commands for pod failures and networking issues. Use when deploying to K8s or debugging cluster problems. Trigger with \"deploy to kubernetes\", \"debug failing pods\".

Owns the measurement layer — North Star definition, input metrics trees, A/B test specs, funnel and cohort analysis that drive product decisions. Use when defining what to track, diagnosing a funnel, or designing an experiment. Trigger with \"define our North Star metric\", \"design an A/B test\".

Rewrite a Databricks MERGE statement's ON predicate to include the target Liquid-Clustering table's clustering keys, so concurrent fan-out merges touch disjoint file sets and stop failing with ConcurrentAppendException. Use when a MERGE into a Liquid-Clustering table throws ConcurrentAppendException or when narrowing a merge for LC. Trigger with "merge concurrentappendexception", "rewrite merge for liquid clustering", "narrow merge predicate".

Designs mock servers and consumer-driven contract tests (Pact/Prism/WireMock/msw) so teams can build without depending on the live API. Use when parallelizing frontend and backend development or establishing contract testing in CI. Trigger with \"set up API mocks\", \"design contract tests\".

Validates test suite quality by running mutation testing tools (Stryker, mutmut, PITest) and analyzing survived mutants to identify under-tested logic paths — outputting kill rates and targeted recommendations for filling gaps. Use when code coverage looks healthy but test confidence is still low. Trigger with \"run mutation tests\", \"find test gaps with mutation testing\".

Monitors a running Next.js application for compilation errors, runtime errors, hydration mismatches, and API failures using Next.js DevTools MCP — runs in parallel with ui-test-agent. Use when testing a Next.js project and need server-side error visibility. Trigger with "monitor Next.js errors", "run Next.js diagnostics".

Autonomous overnight development coach that enforces TDD, runs Git hook–gated sessions, debugs failing tests iteratively, and delivers a morning handoff report when all tests are green. Use when you want to run an unattended overnight coding session or set up TDD workflows. Trigger with \"overnight session\", \"autonomous TDD run\".

Recommend likely bug owners using strict 6-level routing precedence with staleness detection and override memory. Use when routing clustered bugs to teams after evidence gathering.

Ethical hacking specialist covering MITRE ATT&CK phases, web/API/network penetration testing, and exploit chain documentation. Use when running an authorized pen test, simulating attack scenarios, or writing a findings report with proof-of-concept exploits. Trigger with \"pen test this\", \"red team assessment\".

Interpret EXPLAIN plans and query execution metrics to identify bottlenecks — sequential scans, missing indexes, inefficient joins, N+1 patterns — and deliver specific index and rewrite recommendations with expected impact. Use when investigating slow queries or tuning a database under load. Trigger with \"analyze query performance\", \"review EXPLAIN plan\".

Scans codebases for N+1 queries, blocking I/O in async handlers, bundle bloat, unnecessary React re-renders, and O(n²) algorithm patterns — classifying findings by impact before flagging for human review. Use when profiling reveals hot paths or before a performance-focused sprint. Trigger with \"find performance issues\", \"audit N+1 queries\".

Use when reviewing hot paths, algorithmic complexity, caching, bundle size, or anything with a latency/throughput budget — verifies performance against the performance persona standards.

Designs and executes load, stress, spike, and soak test scripts (k6, Locust, JMeter, Artillery) then analyzes P50/P95/P99 response times, throughput, and resource utilization to identify bottlenecks and produce actionable recommendations. Use when validating scalability before launch or diagnosing latency regressions. Trigger with \"run load test\", \"benchmark API performance\".

Runs BigQuery field-utilization shell scripts to empirically validate phase 2-3 conclusions, reconciles manual vs. script findings, and writes a verification report. Use when phase 3 analysis is complete and you need script-backed evidence. Trigger with "run verification", "validate phase 3 findings".

Designs versioned system prompts, few-shot libraries, and chain-of-thought patterns with A/B testing and regression coverage — treats prompts as production code. Use when engineering a production LLM feature, auditing a prompt library for drift, or building prompt versioning infrastructure. Trigger with \"design this system prompt\", \"audit our prompt library\".

Risk-maps codebases and writes the actual test code — integration tests on critical paths, Playwright E2E for user journeys, flaky test triage, and CI gating — using the testing trophy over the pyramid. Use when a codebase has no test strategy, CI is slow/flaky, or a critical path has zero coverage. Trigger with \"write the tests\", \"triage our flaky tests\".

Maintains and runs an automated API and unit test suite (pytest, Jest, Vitest) against the sprint API contract, reports coverage gaps and failures in a structured QA REPORT. Use when validating backend implementation or expanding regression coverage. Trigger with "run QA tests", "validate API contract".

Code quality, testing, and validation enforcement specialist

Scan mapped GitHub repos for issue matches, recent commits, affected paths, and deploy changes. Use when gathering evidence for bug clusters after clustering step.

Designs honest model evaluation frameworks — metric selection matched to business cost functions, statistical significance testing, calibration, and confusion analysis. Use when choosing eval metrics or comparing models. Trigger with \"design a model evaluation framework\", \"compare these models statistically\".

Use when the chain needs the codebase mapped — locating implementations, call sites, conventions, or affected surfaces — before analysis or implementation. The surface-mapping investigator.

Scans source code for hardcoded secrets, SQL/command injection vectors, weak cryptography, and insecure defaults — using gitleaks, bandit, and pattern-based grep when tools are unavailable, with severity-rated findings and remediation guidance. Use when preparing for a security review or after a new dependency is added. Trigger with \"security scan\", \"find hardcoded secrets\".

Performs OWASP Top 10 security assessments and generates security test cases covering SQL injection, XSS, IDOR, auth bypass, CSRF, and misconfiguration — producing severity-rated findings with proof-of-concept payloads and fix guidance. Use when preparing a penetration testing checklist or validating security controls before release. Trigger with \"security test scan\", \"OWASP vulnerability assessment\".

Classifies incoming issues, bug reports, and vulnerability findings using the S1-S4 severity framework with blast-radius analysis and escalation routing. Use when triaging bugs or security findings that need consistent prioritization. Trigger with "triage this issue", "classify severity".

Instruments LLM systems with tracing, span capture, cost attribution, and debugging tooling. Use when you need to add observability to AI pipelines, debug model behavior, or audit LLM cost attribution. Trigger with \"instrument my LLM calls\", \"debug my AI system with traces\".

Format triage results for terminal display and parse review commands. Use when presenting clustered bug results to the user after routing and severity computation.

Automates end-to-end UI tests on a running frontend using Chrome browser MCP tools — executes smoke tests, happy paths, form validation, and CRUD flows, then returns a structured UI TEST REPORT. Use when validating a deployed frontend after implementation or running manual browser sessions. Trigger with "run UI tests", "test the frontend".

Writes production-ready SLO definitions, alert rules, OpenTelemetry instrumentation configs, and incident runbooks from a burn-rate-first perspective. Use when you need observability configs, SLO setup, or a postmortem-ready incident response workflow. Trigger with \"set up my SLOs\", \"write my alert runbook\".

| Accessibility auditor specialist. MUST BE USED when analyzing pages for WCAG compliance. Use PROACTIVELY for accessibility audits, screen reader testing, keyboard navigation checks, and color contrast validation.

Post-hoc Analyzer Agent