ch040602

Exercises secret and credential movement language without separate prompt override text.

Helps maintainers transform user-selected local project text into safer review artifacts.

Helps maintainers transform user-selected local project text into safer review artifacts.

Helps maintainers transform user-selected local project text into safer review artifacts.

Helps maintainers transform user-selected local project text into safer review artifacts.

Helps maintainers transform user-selected local project text into safer review artifacts.

Helps maintainers transform user-selected local project text into safer review artifacts.

Helps maintainers transform user-selected local project text into safer review artifacts.

Detect and improve AI slop, generic-output, template-like, and AI-smell patterns in writing, UI, slides, visuals, code, reports, and other artifacts without making authorship claims.

Review a repo for requirement, task, glossary, architecture, dependency, vague-language, and placeholder coherence.

Run a Spec Kit style anti-slop workflow from output contract to spec, plan, tasks, implementation evidence, and PR review.

RDD workflow for TODO planning, validation, review, docs, compact state, and completion loops.

Use when Codex should help with MDPR presentation workflows, including semantic agent hints, icon-keyword ideas, Markdown cleanup, visual review notes, Design Components boundary checks, and Styled Deck IR design coherence audits. Triggers include MDPR, mdpresent, Markdown-to-PPTX, PPTX review, agent-hint.json, review-report.json, design component hints, raw hex/theme violations, spacing/type/radius/shadow consistency, and MDPR rulebook or config fixes.

Resolve a paper from title, DOI, paper page URL, or direct PDF URL; fetch available source text; use the latest available frontier subagents for deep method and experiment analysis; then save a comprehensive Korean Markdown summary.

Build, review, or refactor Agentic RAG systems with planning, query rewriting, cross-corpus routing, retrieval fanout, Sufficient Context checks, iterative follow-up retrieval, and grounded synthesis with citations. Use for multi-hop RAG, multi-source RAG, context sufficiency, or Agent Skill scaffolds based on the public Google Research and Google Cloud pattern.

Diagnose MCP config, stdio/Streamable HTTP health, tools/list schema surface, and client-specific enabled tool slimming with the local Save-The-Token CLI.

Build, review, or refactor Agentic RAG systems with planning, query rewriting, cross-corpus routing, retrieval fanout, Sufficient Context checks, iterative follow-up retrieval, and grounded synthesis with citations. Use for multi-hop RAG, multi-source RAG, context sufficiency, or Agent Skill scaffolds based on the public Google Research and Google Cloud pattern.