skill
Three-Musketeers
Public Unreviewedby Drew Owens
Execute complex plans and projects with three agents — a Builder, an Adversarial Reviewer, and a Verifier — looping until it genuinely works. The build half of the plan->build pipeline.
Discover
How an AI finds this repo: the auto-discovery snippet from page.json, self-tested against the live discovery search.
Snippet
discovery_triggers
build this with three agentsbuilder adversary verifieradversarial build loopbuild and verify with subagents
discovery_pitch
none (the one-line "when to use this")tags
multi-agentbuildverificationadversarialclaude-codeskill
sample_prompts
Build with the loop
Drive this build with a builder, an adversarial reviewer, and a verifier until it genuinely worksAfter planning
I just finished adom-plan — now build it robustly with three-musketeersAdversarial build
Build this feature, have a separate agent try to poke holes, then verify it as a userVerify by behavior
Build it and verify by actually using it, not by checking against the planInvoke
/three-musketeersSample prompts: self-test
Your own triggers, combined
not in results
How this repo ranks when its full trigger set is the query.
Each sample prompt run against the live discovery search. The score is your rank, does the prompt return this repo at the top?
"Drive this build with a builder, an adversarial reviewer, and a verifier until it genuinely works"Build with the loop
not in results
"I just finished adom-plan — now build it robustly with three-musketeers"After planning
not in results
"Build this feature, have a separate agent try to poke holes, then verify it as a user"Adversarial build
not in results
"Build it and verify by actually using it, not by checking against the plan"Verify by behavior
not in results
"/three-musketeers"Invoke
not in results
No repo matched this query's triggers.