Anvil

Anvil


**Forge the Workflow.** *Structure for vibe coding.*

Anvil is a single, self-contained AI coding agent for the terminal. The coder is a **real agent** — it reads, writes, edits, and runs your project itself (like Claude Code, Cursor, or Aider) — but Anvil wraps it in a **disciplined two-gate review workflow** so a long vibe-coding session can't quietly drift off the rails.

What makes Anvil different isn't the coder — everyone has one of those now. It's the **workflow**: at two human gates, two *different model families* review the work — once on the plan, once on each phase's `git diff` — in a sequential **review → fix → re-review** loop. You approve; the coder does the work; a genuine cross-vendor second opinion keeps it honest.

One Rust binary. Model-agnostic. Installs in one line and updates itself.

**Status: Public Beta.** Anvil is feature-complete for its core workflow and installs/updates cleanly — but it's early. It's been field-tested mostly on **Windows** (macOS/Linux build in CI but haven't been run in anger yet); several big pieces landed recently; and the review-gate flow is verified by hand rather than an automated suite. Expect rough edges, and please [file issues](https://github.com/ai-nhancement/Anvil/issues) — that feedback is exactly what the beta is for. For the best experience, use a capable **tool-calling** model for the coder role.

---

Highlights

• 🔨 Two human gates, cross-vendor review — the plan and each build phase are critiqued by two different model families. Exactly two reviews per gate; no R3+, no drift.
• 🔁 Sequential review loop — R1 reviews → the coder applies fixes → you review → R2 reviews → the coder applies fixes → summary → you approve. R2 deliberately re-reviews after R1's fixes, catching bugs those fixes introduce.
• 🔍 Review anything, not just gates — /review runs the same briefing → critical-review → fix loop on any ad-hoc change that doesn't