FreeSkills
Knowledge wiki setup
- 418418 downloads
- 129129 likes
- 161161 saves
Works with
- Claude Code
- Codex
- Cursor
- and 3 more
Constitutional AI: self-critique and AI feedback v1.0.0
Use this resource to train a model to avoid harmful answers through self-critique
New
- 436 downloads
- 219 likes
- 169 saves
Free
Install with the CLI
In a terminal, from the folder you want:
npx --yes palatekit@latest install constitutional-ai-self-critiqueWith no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.
First time: connect the CLI with
npx --yes palatekit@latest loginThe agent sets up both phases of the method: producing revised pairs through self-critique against a set of principles, then training from AI feedback (RLAIF). You get the pipeline, the principles used and common issues to fix.
Mostly picked by
Updated November 2, 2026
Previews
Details
- Install guides
- Generates with
- Your agent's model
- Licence
- MIT
Similar resources
View all- Free
FreeSkills
Wiki health check
- 552552 downloads
- 202202 likes
- 127127 saves
Works with- Claude Code
- Codex
- Cursor
- and 3 more
FreeSkills
Ask your wiki
- 374374 downloads
- 131131 likes
- 214214 saves
Works with- Claude Code
- Codex
- Cursor
- and 3 more
FreeSkills
Wiki source ingestion
- 315315 downloads
- 230230 likes
- 246246 saves
Works with- Claude Code
- Codex
- Cursor
- and 3 more
FreeSkills
Wiki auditor agent
- 556556 downloads
- 205205 likes
- 192192 saves
Works with- Claude Code
- Codex
- Cursor
- and 3 more
FreeSkills
UX research assistant
- 598598 downloads
- 142142 likes
- 236236 saves
Works with- Claude Code
- Codex
- Cursor
- and 3 more
FreeSkills
Wiki ingestion agent
- 437437 downloads
- 162162 likes
- 191191 saves
Works with- Claude Code
- Codex
- Cursor
- and 3 more
FreeSkills
Wiki librarian agent
- 304304 downloads
- 120120 likes
- 217217 saves
Works with- Claude Code
- Codex
- Cursor
- and 3 more
What you get
- SKILL.md: quick start and workflows for the supervised phase
- SKILL.md: RLAIF phase, AI preferences and chain-of-thought critique
- SKILL.md: when to use the method and common issues
Good fit for
- You want to reduce harmful answers without labeling thousands of examples
- You are studying the alignment method described by Anthropic
Not for
Tags
- constitutional ai
- alignment
- rlaif
- safety
About
You say
“Write a five-principle constitution and generate revised response pairs for my model.”
Skill published by davila7 under the MIT license, delivered in its original version.
Show moreShow less
How to use it
- Install the skill in your agent's skills folder
- Ask your agent to train a model to avoid harmful answers through self-critique
- Review and adjust the result
Content preview
Files
- LICENSE1.1 KB
- NOTICE-PALATEKIT.md0.5 KB
- SKILL.md8.3 KB
Origin and licence
Listed unmodified
- Author
- davila7
- Source
- github.com/davila7/claude-code-templates/tree/f57964ebda91cdfb31755661ecea9395f30020f1/cli-tool/components/skills/ai-research/safety-alignment-constitutional-ai
- Licence
- MIT
Notice
« safety-alignment-constitutional-ai » (davila7, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.
User reviews
–
No ratings yet
Loading reviews…