discernment-nudge
skills/discernment-nudge · anthropics/skills · by Anthropic
After you give a substantive answer or draft that the user may act on · advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument · invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review · the skill file explains these boundaries and the exact output format.
Source · GitHub API and the SKILL.md frontmatter, read 2026-11-15 · stars are for the whole anthropics/skills repository
Install discernment-nudge
Ready-to-paste · from the repository README and the skill folder
# Copy this skill folder into your personal Claude Code skills git clone --depth 1 --filter=blob:none --sparse --branch main https://github.com/anthropics/skills.git /tmp/discernment-nudge-src git -C /tmp/discernment-nudge-src sparse-checkout set skills/discernment-nudge mkdir -p ~/.claude/skills cp -R /tmp/discernment-nudge-src/skills/discernment-nudge ~/.claude/skills/discernment-nudge
Anthropic's public skills repository, anthropics/skills · each folder holds a SKILL.md with a name and description in its YAML frontmatter. Claude Code reads personal skills from ~/.claude/skills/<name>/SKILL.md.
Alternatives
Same category · other repositories · latest commit first
More from Anthropic
Same repository · anthropics/skills · latest commit first
Embeddable badge
Drop into your README · links to this skill page
[](https://benchgecko.ai/skills/claude-discernment-nudge)
FAQ
Built from the GitHub data above
What is discernment-nudge?
discernment-nudge is a Claude Skills agent skill by Anthropic, in the skills/discernment-nudge folder of the anthropics/skills repository (BenchGecko category: Other). Its SKILL.md describes it as: After you give a substantive answer or draft that the user may act on · advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument · invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review · the skill file explains these boundaries and the exact output format.
Where does this data come from?
From GitHub, read 2026-11-15: the name and description from the SKILL.md frontmatter, 2 files from the repository tree, the last commit to the folder on 2026-08-17, and 195,006 stars for the whole anthropics/skills repository. BenchGecko does not rate skills.
How do I install discernment-nudge?
Copy the skills/discernment-nudge folder into ~/.claude/skills/ for Claude Code, use the command from the anthropics/skills README, or point any client that supports the Agent Skills format at the folder. Ready-to-paste snippets are above.
Is discernment-nudge official?
discernment-nudge is published in anthropics/skills, a repository owned by the Anthropic organization on GitHub.
See also
Keep exploring the BenchGecko graph