ClaudeMap

HERO Anti-OverDefense

A paste-in RULES.md contract that stops coding agents from over-defending their changes.

Prompt Librariespromptingrulesagent-behaviorconfig
Visit projectwanshuiyin/HERO-Anti-OverDefense

HERO targets four over-defense behaviours coding agents drift into — Hashing, Edge cases, Rubrics, Overbuild — where the agent optimises for not being blamed rather than for the work being good. Eight rules bound proposed fixes without suppressing legitimate work; ships with a quoted case catalogue and calibration examples of what is not over-defense. Nothing to install: paste into CLAUDE.md, AGENTS.md, or any agent's config file.

Repository

wanshuiyin/HERO-Anti-OverDefense

Charted

Related on the map

  • prompts.chat

    Prompt Library

    The de-facto standard ChatGPT role-prompt collection (f.k.a. Awesome ChatGPT Prompts).

    • prompts
    • awesome-list
    • roles
    • community
    View

  • Claude Quickstarts

    Prompt Library

    Anthropic's official Claude API quickstarts — agentic customer service, RAG, skills, and more, in working notebooks.

    • official
    • anthropic
    • quickstart
    • examples
    View