Item detail

ksimback/looper

Looper is an MIT-licensed Claude Code skill and slash command for designing review-gated agent loops before execution, with loop interviews, ASCII and Mermaid previews, reviewer or judge gates, stop conditions, portable loop specs, and an optional same-session handoff runner.

Score8.3
Popularity52.0
Riskconditional
TierSilver
Score breakdown
Usefulness8.0
Novelty8.0
Momentum5.0
Maturity7.1
Open-source/build8.4
Evidence7.2
Workflow potential9.8
Setup ease6.4

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for people who keep improvising agent loops on the fly and want a more deliberate way to define goal, verification, review gates, and stop conditions before a coding agent starts changing files.

Who should use it

Claude Code usersdevelopers designing agent workflowsteams using reviewer or judge modelspeople standardizing repeatable coding loops

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

Looper can hand project context to a different reviewer model or vendor CLI and can write loop artifacts into the current workspace, so test it first on a non-sensitive repo and inspect what gets shared.

Evidence links

Closest alternatives / related signals

claude-codeagent-loopsworkflow-designreview-gatesskills