AnthropicClaude Sonnet 5 and Claude Opus 5

The Claude 5 learning curve

Give Claude a clear outcome and the right project context, then remove legacy scaffolding that newer models no longer need.

Start with the promptReview the evidence

The mental model

What changes the result

“Less instruction can produce more capability when the remaining context is high-signal.”

Three levels

Climb the curve in this order

Do not add production scaffolding before you can describe and measure a good single-task result.

  1. 01

    Start with the outcome

    Get a strong first result without teaching Claude a complicated ritual.

    • State the artifact or change you want, the relevant context, and what “done” means.
    • Use direct, positive instructions for tone, length, and scope.
    • For simple work, begin with the default behavior before adding special rules.
    Ready to move on when: A new teammate can read the request and predict the expected deliverable.
  2. 02

    Calibrate effort and examples

    Make repeated outputs predictable without over-constraining the model.

    • Use a representative good example when style or format matters.
    • Raise effort for genuinely hard work; do not simulate deeper reasoning with long step lists.
    • Describe scope explicitly because newer Claude models follow literal boundaries closely.
    Ready to move on when: The prompt succeeds on several real cases, not only the example used to write it.
  3. 03

    Engineer the whole context

    Keep long-running agent work focused as tools, memory, and instructions accumulate.

    • Keep CLAUDE.md focused on durable project truth; move procedures into on-demand Skills.
    • Make tool names, descriptions, inputs, and return values precise.
    • Review old anti-laziness, narration, and verification instructions after every model upgrade.
    Ready to move on when: Removing an instruction does not reduce eval quality, or the instruction earns its place.

Starter template

Begin here, then earn every extra instruction

Replace the bracketed text, run it on a real task, and add guidance only when you can name the failure it fixes.

claude-5-starter.txt
# Outcome
Implement [specific result].

# Project context
- Relevant files or systems: [paths / links]
- Existing pattern to follow: [example]

# Boundaries
- Do not change: [out-of-scope areas]
- Ask before: [meaningful decisions]

# Done when
- [observable acceptance check]
- [test or verification]

Work autonomously inside these boundaries. Prefer the simplest complete solution.

Patterns that work

Progressive disclosure

Load specialized procedures only when the task calls for them instead of placing every rule in the system prompt.

General heuristics

Describe the decision boundary and desired result; allow the model to choose implementation steps.

Explicit deliverable scope

Say whether you want a patch, analysis, document, or recommendation and how complete it must be.

Common failure modes

The inherited mega-prompt

Claude over-explains, over-verifies, or follows obsolete rituals.

Fix: Delete rules that compensate for older model behavior and retest from a minimal baseline.

Prescribing every reasoning step

The model follows a weaker human-authored plan even when it can find a better path.

Fix: Specify the goal, constraints, and checks; prescribe steps only when the process itself is required.

Always-on procedures

Unrelated instructions consume attention and trigger in the wrong tasks.

Fix: Keep project facts in CLAUDE.md and package detailed workflows as discoverable Skills.

Current as of July 26, 2026

What the latest guidance changes

The Claude Code team’s July 2026 field note says it removed roughly 80% of its system prompt for the newest models. The durable lesson is not “short prompts win”; it is that every context token should justify its effect.

Evidence ledger

Read the sources behind the advice

Official provider documentation takes priority. Maintainer field notes are useful evidence, but they are labeled separately.

Claude prompting best practices

official

Anthropic

Current model behavior, adaptive thinking, examples, scope, and migration guidance.

Prompting Claude Sonnet 5

official

Anthropic

Effort calibration, literal instruction following, verbosity, and thinking defaults.

The new rules of context engineering

maintainer

Thariq Shaukat, Claude Code

The field note that prompted this guide and the reported system-prompt reduction.

This guide is reviewed at least every 30 days. Automated checks flag unavailable sources and overdue reviews; substantive advice changes require human verification.