Skill · em Dados, IA e pesquisa

constitutional-ai

Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.

Procedência

Antes de instalar

1 arquivo · 8 KB · só texto, nenhum script

Instalar na sua CLI

O comando baixa a versão fixada (commit 57f899e) direto da origem, para a pasta que a CLI lê. Precisa de curl (macOS e Linux); no Windows não há comando, porque o Rook Labs é para macOS.

Claude Code

Neste projeto: instala em .claude/skills/constitutional-ai/.

d=".claude/skills/constitutional-ai"
u="https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/skills/ai-research/safety-alignment-constitutional-ai"
curl -fsSL --create-dirs \
  -o "$d/SKILL.md" "$u/SKILL.md"

Global: instala em ~/.claude/skills/constitutional-ai/.

d="$HOME/.claude/skills/constitutional-ai"
u="https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/skills/ai-research/safety-alignment-constitutional-ai"
curl -fsSL --create-dirs \
  -o "$d/SKILL.md" "$u/SKILL.md"

Codex

Neste projeto: instala em .agents/skills/constitutional-ai/.

d=".agents/skills/constitutional-ai"
u="https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/skills/ai-research/safety-alignment-constitutional-ai"
curl -fsSL --create-dirs \
  -o "$d/SKILL.md" "$u/SKILL.md"

Global: instala em ~/.agents/skills/constitutional-ai/.

d="$HOME/.agents/skills/constitutional-ai"
u="https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/skills/ai-research/safety-alignment-constitutional-ai"
curl -fsSL --create-dirs \
  -o "$d/SKILL.md" "$u/SKILL.md"

Antigravity

Neste projeto: instala em .agents/skills/constitutional-ai/.

d=".agents/skills/constitutional-ai"
u="https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/skills/ai-research/safety-alignment-constitutional-ai"
curl -fsSL --create-dirs \
  -o "$d/SKILL.md" "$u/SKILL.md"

Global: instala em ~/.gemini/antigravity-cli/skills/constitutional-ai/.

d="$HOME/.gemini/antigravity-cli/skills/constitutional-ai"
u="https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/skills/ai-research/safety-alignment-constitutional-ai"
curl -fsSL --create-dirs \
  -o "$d/SKILL.md" "$u/SKILL.md"

Peça ao Rook

Já usa o Rook Labs? Cole no chat do Rook: instale a skill https://rooklabs.sh/marketplace/cct.constitutional-ai

Prévia do SKILL.md

---
name: constitutional-ai
description: Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.
version: 1.0.0
author: Orchestra Research
license: MIT
tags: [Safety Alignment, Constitutional AI, RLAIF, Self-Critique, Harmlessness, Anthropic, AI Safety, RL From AI Feedback, Claude]
dependencies: [transformers, torch, trl]
---

# Constitutional AI - Harmlessness from AI Feedback

## Quick start

Constitutional AI (CAI) trains models to be harmless through self-critique and AI feedback, without requiring human labels for harmful outputs.

**Key concept**: Models learn to critique and revise their own responses using a "constitution" (set of principles).

**Two phases**:
1. **Supervised Learning (SL)**: Self-critique + revision
2. **Reinforcement Learning (RL)**: RLAIF (RL from AI Feedback)

**Constitution example**:
```
Principles:
1. Choose the response that is most helpful, honest, and harmless
2. Avoid responses that are toxic, racist, or sexist
3. Prefer responses that explain objections rather than refuse
4. Choose responses that are thoughtful and nuanced
```

## Common workflows

### Workflow 1: Supervised learning phase (self-critique + revision)

**Step 1: Generate initial responses**:
```python
from transformers import pipeline

generator = pipeline("text-generation", model="base-model")
…

Ver todo o marketplace