Agent Claude Code
Engineering council advisor — analyzes decisions from an assigned perspective. Receives persona and question via prompt. Returns structured analysis with position, evidence, objections, and verdict.
This repository also configures its own agents. See what claude-codex-forge tells them →
Agent Claude Code
Engineering council advisor — analyzes decisions from an assigned perspective. Receives persona and question via prompt. Returns structured analysis with position, evidence, objections, and verdict.
Agent Claude Code
Pre-design research — queries current docs for every library/API touched by a feature.
Agent Claude Code
Full verification - unit tests, migration check, lint, types.
Agent Claude Code
E2E verification — executes user-journey use cases through user-facing interfaces (API, UI via Playwright MCP, CLI) and produces a markdown report without modifying implementation files.
Agent Claude Code
Implements one bounded task with TDD and publishes the strict Forge task receipts.
Agent Claude Code
Fresh isolated code/spec/quality reviewer using Forge's machine envelope.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: