Diagnostician — Testing & QA agent for Claude Code
Finds the cause of a failing check, test or CI run in a devflow task and reports it with evidence, without changing files.
How to install Diagnostician
Installs to ~/.claude/agents/wotjr1649-devflow-diagnostician.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/wotjr1649/devflow/HEAD/agents/diagnostician.md -o ~/.claude/agents/wotjr1649-devflow-diagnostician.md Restart Claude Code, or start a new session, for it to be picked up.
What Diagnostician does
name: diagnostician description: Finds the cause of a failing check, test or CI run in a devflow task and reports it with evidence, without changing files. Use when a failure's cause is unclear, before anyone fixes it. model: sonnet effort: high tools: Read, Grep, Glob, Bash, PowerShell
You find why something fails; someone else fixes it. Keeping the two apart is the point: a fix written while the cause is still a guess tends to cover the symptom.
Reproduce the failure first and say ho
Alternatives in Testing & QA
- Grid Tester — Use this agent for test strategy, TDD workflows, integration/e2e coverage, and flaky test hardening 523 ★
- Codex Debugger — Root-cause a failing test, crash, stack trace, or misbehaving feature by delegating the investigation to OpenA 249 ★
- Eval Improver — Improver role of the D365FO agent eval loop 136 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Test Triager
Diagnoses failing tests, builds, type errors, and CI failures — reproduces the failure, finds the first meanin
Test Maintainer
Fixes failing and flaky tests, and refactors a suite that has grown duplication. Finds the root cause instead
Auto Fixer.Agent
Use when there is a failing test, error log, CI failure, review finding, production symptom, or other concrete
Flake Investigator
Establishes why a Playwright case failed intermittently and classifies the cause before anyone changes the tes
Claudehut Test Runner
Runs the test suite and reports real pass/fail counts and failure causes — the fresh test evidence Review reco
Task Builder
Implement a frozen task spec and later package evidence without changing production code during evidence mode
Related Skills
Logic Locate
Locate the root cause of a confirmed failure — use when you have a stack trace, failing test, or wrong output
St Debug
Hypothesis-driven debugging with observation-only instrumentation and user reproduction; fixes route through t
GitHub Review Failures
Use when CI checks are failing on a PR — fetches failure logs, diagnoses root causes, implements fixes, and pu
