Toby — Testing & QA agent for Claude Code
Write failing tests from an approved plan, before any implementation exists.
How to install Toby
Installs to ~/.claude/agents/namandembla-agentic-sdlc-toby.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/NamanDembla/agentic-sdlc/HEAD/agents/toby.md -o ~/.claude/agents/namandembla-agentic-sdlc-toby.md Restart Claude Code, or start a new session, for it to be picked up.
What Toby does
name: toby description: Write failing tests from an approved plan, before any implementation exists. Use after the user has approved the plan at the plan gate, and again when Kevin reports that a test itself is wrong rather than the code. Dispatched by Michael at the tests stage. Writes test files and tests.md mapping each test to an acceptance criterion. Never writes production code. tools: Read, Grep, Glob, Bash, Write, Edit model: sonnet color: green
You are Toby. You are the one who
Alternatives in Testing & QA
- Rivetkit Driver Test Writer — Use this agent when the user needs to write, modify, or debug driver tests for RivetKit 6.1k ★
- Grid Tester — Use this agent for test strategy, TDD workflows, integration/e2e coverage, and flaky test hardening 523 ★
- Propagate — Generate tests from Allium specifications 474 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Sim Engineer
Use to write or fix sim.js, the pure simulation logic for the Monty Hall problem (3-door and N-door), after SP
Merge Gate Verifier
Read-only verification of a branch before it merges. Use when a worker reports a lane done and you need to che
Ae Test
Agent Engineer test agent. Writes tests from the acceptance criteria rather than from the implementation, muta
Spec Checker
Compares the implementation against the spec file named in PLAN.md and PLAN.md itself, and reports every requi
Test Developer
Writes failing tests from a story's acceptance criteria before any production code exists (the RED phase). Use
Spec Challenger
Adversarial reader of a draft spec, invoked by /architect before the GitHub issue is created. Starts with clea
Related Skills
Thumbgate Guard
Turn the last agent mistake into a hard prevention rule the agent cannot bypass. Use after a bad tool call, a
Clear Explain
Explain one CLEAR-100 rule: what it catches, why it exists, the fix, and a before-and-after. Use when a findin
Evolve Voice
Run one evolution round: mass-generate the format matrix, adversarial judge panels, master board, empirical A/