Anthropic: Claude prompting best practices
Anthropic's main reference for prompting current Claude models: clear instructions, examples, structure, tools, thinking and agents.
- Kind
- Prompt
- Works with
- Any agent, Claude Code
- Source
- Official prompting guides (Claude, OpenAI, Google, media tools) @ 18a1eaf5414e
- Path
- items/anthropic-prompting-best-practices
- Jobs
- Documentation
- Topics
- Prompt engineering
- Tags
- prompting-guide, official
Credit
- Made by
- Skill Manifest curation
- Original
- platform.claude.com/docs/en/build-with-claude/prompt-engineer…
- License
- No licence found
Listed by Skill Manifest; not made by us.
InstallCopy it from the source repository, or sign in and let the assistant adapt it for your project.
Install factsScanned 2 Nov 2026
- Adds
- A prompt to paste or save as a command
- Runs scripts
- None found
- Uses the network
- None found
- Needs secrets
- None found
- Source revision
- 18a1eaf5414e
- Licence
- Unknown
- Last activity
- Unknown
- Review
- Reviewed
Found by a static scan of the stored text; nothing is run. “None found” is not a guarantee, so read what you install before you run it.
- The source states no licence
- The source's last activity is not known yet
# Anthropic: Claude prompting best practices Anthropic's main reference for prompting current Claude models: clear instructions, examples, structure, tools, thinking and agents. Key rules: be clear and direct; add the context the model needs; give a few good examples; structure long prompts with XML-style tags; give the model a role; control output format and length; guide tool use and thinking; for agents, keep scope tight to avoid over-eager changes. Covers Claude Fable 5.1, Mythos 5.1, Opus 5.5, Sonnet 5.5, Haiku 5.5 and some older 4.x models, with a page for each. Checked 2 Nov 2026. The old prompt-library address now redirects to this page, so no separate prompt-library page exists. URL: https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices