# Spec, TDD and review for coding agents

> For developers and tech leads who want coding agents to work like careful engineers: guardrails first, then interview, spec, test-first build, debug, verify, review. Most run in Claude Code or Codex; the three Anthropic plugins are Claude Code only. Agent review assists; a person approves merges.

- Page: https://hebbmind.com/collections/spec-tdd-review-coding-agents
- Updated: 2026-10-12

## Resources

- [CC Safety Net](https://hebbmind.com/plugin/cc-safety-net.md): [Plugin] Set up first: blocks git reset --hard, rm -rf on risky paths, reads of .env or SSH keys. Node 18+; not a sandbox.
- [Anthropic Sandbox Runtime (srt)](https://hebbmind.com/cli/sandbox-runtime.md): [CLI] Set up first: Anthropic's OS-level file and network limits. Reads open by default; Linux needs bubblewrap; preview.
- [Security Guidance](https://hebbmind.com/plugin/security-guidance.md): [Plugin] Always on: flags risky edits; Opus reviews diffs and commits (extra usage). Claude Code 2.1.144+, Python 3.8+.
- [Grill Me](https://hebbmind.com/skill/grill-me.md): [Skill] Step 1: the agent interviews you about the plan until every decision is settled. You start it; install grilling too.
- [Spec Kit](https://hebbmind.com/cli/spec-kit.md): [CLI] Step 2: GitHub's toolkit turns your answers into a spec, plan and tasks, then builds against them. Python 3.11+ and uv.
- [Context7](https://hebbmind.com/mcp/context7.md): [MCP] Step 3: feed the agent current, version-specific library docs. Upstash service; a free API key raises limits.
- [Test-Driven Development](https://hebbmind.com/skill/test-driven-development.md): [Skill] Step 4: strict red-green-refactor; code written before a failing test is deleted. Also in the Superpowers plugin.
- [Diagnosing Bugs](https://hebbmind.com/skill/diagnosing-bugs.md): [Skill] Step 5, when it breaks: repro loop, hypotheses, fix, then a regression test. Redacts credentials in output.
- [Verification Before Completion](https://hebbmind.com/skill/verification-before-completion.md): [Skill] Step 6: the agent must run the full checks and read the output before it claims it is done. Also in Superpowers.
- [Code Review](https://hebbmind.com/plugin/code-review.md): [Plugin] Step 7: four parallel agents review a GitHub PR and keep issues scored 80+. Claude Code only; needs a logged-in gh.
- [PR Review Toolkit](https://hebbmind.com/plugin/pr-review-toolkit.md): [Plugin] Step 7: six reviewers for tests, silent failures, comments, type design, style and simplification. Claude Code only.
- [Supply Chain Risk Auditor](https://hebbmind.com/skill/supply-chain-risk-auditor.md): [Skill] Before merge: flags advisories, abandoned repos, install scripts in npm, PyPI, Go deps. Needs uv; skips pnpm/yarn locks.

[Collections](https://hebbmind.com/collections.md)