Devloop, a local code/review loop for Codex and Claude Code
Adversarial Codex/Claude setup prevents agents from favoring their own outputs.
Run Codex app-server reviews of uncommitted changes via an MCP tool.
Wraps Codex /review in MCP to separate review from implementation context.
Developers using Codex CLI for code review
Codex CLI · Cursor · GitHub Copilot
This works by exposing to agent single new mcp tool call allowing it to request review.
MCP creates new codex instance in the background and uses app-protocol to request identical flow as /review with option to include additional review instructions.
GPT-5.5 started to be very good and providing meaningful instructions, describing what was desired intent of introduced changes.
This works much better than /review + "fix it" approach, as review result is passed to agent that have while implementation knowledge.
This resolves 2 failure modes:
1. Review asks to restore what we are trying to change
2. Flapping between 2 issues (first review resolves issue by introducing second one, second issue reintroduces first issue)
I decided to post this mcp right now, as it looks like there is community interest for this particular solution, therefore I wanted to present my existing and already battle tested implementation.
Adversarial Codex/Claude setup prevents agents from favoring their own outputs.
Terminal-first AI code review with MCP integration for agent feedback loops.
Automated code review loop via agent ping-pong, but Cursor already does multi-turn fixing in context.
Orchestrates two AI tools you already pay for; requires both Codex and Claude.
Browser review layer for terminal agents when most tools are CLI-only.
Work-review-gate loop automates Claude iteration, but templated agentic loops already exist at scale.