01 / System
Agent Factory Review Gate
Keep AI-agent output moving without letting unreviewed changes leak into public surfaces.
A bounded handoff and review system for dispatching implementation work, preserving scope, and requiring proof before public action.
Loop Map
Lock scope, boundaries, and route contracts before implementation.
Let the code worker rebuild or patch inside the assigned worktree.
Run lint, build, route smoke tests, and visual QA where relevant.
Capture known gaps and follow-up issues before review.
Hold push, deploy, publishing, Search Console, and billing actions for human approval unless explicitly opened.
Protocol
The operating loop.
A bounded handoff and review system for dispatching implementation work, preserving scope, and requiring proof before public action.
- 01Lock scope, boundaries, and route contracts before implementation.
- 02Let the code worker rebuild or patch inside the assigned worktree.
- 03Run lint, build, route smoke tests, and visual QA where relevant.
- 04Capture known gaps and follow-up issues before review.
- 05Hold push, deploy, publishing, Search Console, and billing actions for human approval unless explicitly opened.
Objective Dispatcher Pattern
Route tasks across agents while preserving the core objective.
Sandbox Execution Pattern
Run unsafe actions inside controlled execution sandboxes.
Coding Agent Wrappers: Convenience, Durability, and Policy Risk Without the Hype
A practical guide to coding agent wrappers: where they help, where they degrade workflow quality, and how to judge native CLI vs wrapper vs harness without policy melodrama.
Cross-Agent Handoff: How to Move Work Between Coding Agents Without Losing Continuity
A practical field guide to cross-agent handoff: what belongs in the packet, when to resume instead of switch, and how to move work between coding agents without turning the workflow into mush.
OpenAI Symphony Review: What It Actually Does
OpenAI Symphony is an issue-driven coding-agent orchestrator with repo-owned workflow contracts, reconciliation loops, and per-issue workspaces. Here’s what it actually does and what builders should steal.
Output
A reviewer-ready branch with evidence, not a pile of local changes and assumptions.