Converts specification documents into phase-driven checklists with enforcement gates.
npx skills add https://github.com/lukacf/meerkat --skill rct-methodology
Please help me install this Agent Skill. Skill: rct-methodology Source: https://github.com/lukacf/meerkat/tree/main/.claude/skills/rct-methodology Command: npx skills add https://github.com/lukacf/meerkat --skill rct-methodology
Automates AI development governance by turning specifications into executable, phase-driven checklists and gates, ensuring representation contracts are validated before implementation.
Scaffold the Luka Loop in your repository and start the gated implementation workflow.
High-intent search queries and answers about installing and using this skill.
To convert a specification into a gated checklist, this methodology creates .rct/spec.yaml, .rct/plan.yaml, and .rct/checklist.yaml files, enforcing phase-driven gates and representation contracts before implementation.
Phase-driven gates are strict checkpoints (Gate 0–Gate 3) that validate specification compliance. They use independent reviewer prompts to ensure representation contracts are met before behavior implementation proceeds.
To prevent must-have feature deferral, you apply strict governance checklists that validate completion via rendered outputs in .rct/outputs/CHECKLIST.md, ensuring all specified features pass gate reviews without deferral.
Yes, the RCT methodology requires independent reviewer prompts to run gate reviews. This ensures auditable process enforcement and validates that representation contracts are satisfied before implementation behavior.
Yes, you can use gated checklists to maintain artifact isolation by storing all governance files within the .rct/ directory, preventing spec deferral and rendering progress outputs cleanly to .rct/outputs/.
You should not use a strict checklist methodology for rapid prototyping or exploratory development where spec requirements fluctuate, as the phase-driven gates and representation contract enforcement demand fixed specifications.
ormastesDefine ownership boundaries for host state and shared LLM process assets.
futuregeraldDispatch fresh subagents per task to execute implementation plans with two-stage reviews.
fredmilhomeDraft structured preregistration documents for research studies in OSF, AsPredicted, or AEA RCT Registry styles.
robertbr123Automate task routing and assignment across Kanban boards to specialists.
