Execute iterative PLAN, PRODUCE, EVALUATE, CRITIQUE, REFINE loops with quality gates.
npx skills add https://github.com/drsapaev/final --skill aif-loop-drsapaev
Please help me install this Agent Skill. Skill: aif-loop Source: https://github.com/drsapaev/final/tree/main/.codex/skills/aif-loop Command: npx skills add https://github.com/drsapaev/final --skill aif-loop-drsapaev
This Skill automates complex, multi-step refinement processes for artifacts like code, documentation, or specifications, ensuring they meet predefined quality standards through iterative cycles.
Use the aif-loop skill to start a new loop for generating an OpenAPI 3.1 specification based on DDD principles.
High-intent search queries and answers about installing and using this skill.
Iterative refinement of artifacts uses a strict multi-phase loop with PLAN, PRODUCE, EVALUATE, CRITIQUE, and REFINE stages to improve output until quality gates are met or iteration limits are reached.
Quality gates in artifact generation are predefined criteria that outputs must pass before completion, ensuring the LLM agent prevents low-quality results through structured evaluation and iterative refinement cycles.
Automating generate-critique-refine workflows involves executing a multi-iteration loop that produces an artifact, evaluates it against defined rules, critiques flaws, and refines the output until correctness requirements are satisfied.
Iterative refinement supports complex documentation and code generation by automating multi-step processes, evaluating outputs against quality standards, and refining artifacts through structured cycles until completeness is achieved.
When an artifact fails quality gates during the AI loop, the system critiques the identified flaws and triggers another refinement iteration, continuing until criteria are met or maximum iteration limits are reached.
The aif-loop skill operates without external dependencies, using internal references to execute iterative refinement loops for quality-gated artifact generation without requiring additional environment setup.
joshsymondsCreate and modify Claude skills with evaluation-driven testing cycles.
thec3ukEnforce centralized testing standards when writing test artifacts.
viktorbezdekDiagnose LLM spending and apply context, prompt, and model optimizations with a quality gate.
dirienDetect linting and validation failures and fix them atomically across project toolchains.
