Panel review loop

SkillDev tools

Lets your agent iteratively review and improve a user-facing surface across real-world projects until a review panel has no blocking concerns.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Panel review loop skill

About this capability

Iteratively review and improve a Fallow user-facing surface across representative real-world projects until the panel has no blocking concerns.

What this skill tells your AI

The instructions your AI receives, as published by fallow-rs/fallow in .agents/skills/panel-review-loop/SKILL.md and read by ahel’s review.

The real-world corpus is intentionally untracked. Before its first use, follow the benchmark setup and run npm --prefix benchmarks run download-fixtures.

  1. Define the user-visible surface and success criteria.
  2. Select representative projects from benchmarks/fixtures/real-world/.
  3. Capture actual output for each project.
  4. Run panel-review on the evidence, not on an intended answer.
  5. Implement the smallest coherent improvement that addresses consensus.
  6. Re-run the same corpus and compare behavior.
  7. Stop when the panel has no blocks and further changes do not materially improve the surface.

Keep the corpus stable across iterations. Preserve output contracts unless the plan explicitly approves a versioned change.

Signals

GitHub stars
4k
Forks
155
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
panel-review-loop
Source
github.com/fallow-rs/fallow