Kiterlin/anti-defensive-writing

Codex skill for removing defensive writing and strengthening prose.

What it solves

This project addresses "defensive writing" in academic and professional prose—the habit of over-protecting arguments with excessive hedging, preemptive apologies, and self-limiting disclaimers that dilute the impact of core insights.

How it works

It provides an open agent skill and prompt standard that guides AI agents to audit and rewrite hesitant drafts. The system identifies patterns like stacked modal hedging and negative framing, then transforms them into direct, claim-forward prose while preserving necessary scientific precision and methodological constraints.

Who it’s for

Researchers, academics, and professionals who write research papers, grant proposals, and professional briefs and want to present their findings with more authority and precision.

Highlights

  • Prompt Standard: Includes specific system prompts for ChatGPT, Claude, and AI code editors like Cursor and Windsurf.
  • Agentic Workflow: A three-step process of auditing, reviewing, and revising to ensure technical accuracy is not lost during the rewrite.
  • Multi-platform Support: Installation scripts for CLI agents and configuration files for various LLM interfaces.
  • Case Studies: Real-world before-and-after examples for academic introductions and grant proposals.

Related

  • Project
    Azure-Samples/rag-postgres-openai-python

    A web-based chat application that uses RAG to answer questions about PostgreSQL database rows using hybrid search and OpenAI function calling.

  • Project
    oOo0oOo/lean-lsp-mcp

    An MCP server that enables LLM agents to interact with the Lean theorem prover via LSP, providing tools for proof state analysis and theorem searching.

  • Project
    foxminchan/BookWorm

    A cloud-native reference application demonstrating multi-agent AI orchestration and standardized AI tooling integrated into a microservices architecture.

  • Project
    aigc-apps/PAI-RAG

    An open-source Agentic RAG framework that combines multimodal understanding, ReAct-based agents, and enterprise-grade knowledge management for complex intelligent Q&A.