Harm Anticipation
Proactively identifying failure modes, misuse, and unintended consequences.
Install
/plugin marketplace add Owl-Listener/ai-design-skillsThe repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.
npx skills add Owl-Listener/ai-design-skills --skill harm-anticipationgit clone --depth 1 https://github.com/Owl-Listener/ai-design-skills.git /tmp/ai-design-skills
mkdir -p ~/.claude/skills
cp -r /tmp/ai-design-skills/claude-plugin/ai-alignment-reasoning/skills/harm-anticipation ~/.claude/skills/harm-anticipationPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Harm Anticipation Harm anticipation is systematically thinking through how an AI product could cause harm — before it does. Preventive design, not reactive crisis management. The work is unglamorous and easy to skip. Done well, it produces specific testable mitigations. Done badly, it produces a doc nobody reads. ## Categories of AI harm - **Direct harm**: the AI outputs something harmful — dangerous advice, discriminatory content, privacy violations - **Facilitated harm**: the AI helps a user do something harmful, even if the AI's output itself is benign - **Emergent harm**: harmful patterns from scale or interaction effects, not from any single output - **Omission harm**: the AI fails to act when it should — not flagging a crisis, not escalating - **Erosion harm**: gradual negative effects — dependency, deskilling, manipulation, trust erosion ## Structured anticipation Work through each harm category with five lenses: 1. **Who could be harmed?** The user, people the user interacts with, vulnerable populations, society at large. 2. **How could they be harmed?** Physical, emotional, financial, reputational, privacy, autonomy. 3. **What's the likelihood?** Common use case vs. edge case vs. adversarial attack. 4. **What's the severity?** Inconvenience vs. distress vs. irreversible damage. 5. **What's the detectability?** Obvious and immediate vs. subtle and delayed. ## Misuse scenarios Think like an adversary: - How would someone deliberately misuse this feature?
Details
- Skill name
harm-anticipation- Repository
- Owl-Listener/ai-design-skills
- Path
claude-plugin/ai-alignment-reasoning/skills/harm-anticipation/SKILL.md- License
- MIT
- Repo updated
- 9 June 2026
FAQ
What does the Harm Anticipation skill do?
Proactively identifying failure modes, misuse, and unintended consequences.
How do I install the Harm Anticipation skill?
Claude Code plugin: /plugin marketplace add Owl-Listener/ai-design-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Harm Anticipation popular?
It lives in Owl-Listener/ai-design-skills, which has 207 GitHub stars and 37 forks.
Related skills
Agent Role Design
207Owl-Listener/ai-design-skills
Defining what each agent does, knows, and owns in a multi-agent system.
Behavioral Consistency
207Owl-Listener/ai-design-skills
Ensuring the AI behaves predictably across sessions, edge cases, and modalities.
Bias Detection Design
207Owl-Listener/ai-design-skills
Designing review workflows to surface and mitigate bias in AI outputs.
Publish Package
292danielvm-git/bigpowers
Package registry publishing for npm, crates.io, PyPI, and Homebrew. Verifies prerequisites, runs the publish command, confirms success, and surfaces actionable error hints on failure.
Frustration Detection
207Owl-Listener/ai-design-skills
Reading user emotional state from text signals — caps, punctuation density, repetition, latency — and adapting before the user disengages.
Session State
292danielvm-git/bigpowers
Track implementation decisions and progress in specs/state.yaml to prevent context rot. Use at the start of a session to load context, and whenever a significant decision is made or a milestone is reached.
Browse the most popular Claude skills, all other skills, or write your own with the skills generator.