CClaude Weekly DigestIdeas for building thoughtfully with AI
Browse studies

A guided way into the collection

Start with the
question you have.

You do not need to read chronologically. Choose the path closest to your work, follow three or four connected studies, then use the full collection when you want to go deeper.

01Build a stronger foundation02Design agent workflows03Make implementation reliable
Path 01

Build a stronger foundation

Begin with the habits that make Claude easier to direct: clarity, context, examples, and choosing the right control surface.

  1. 01
    Claude Prompting Study · 4 min

    Use Examples Effectively

    Examples are one of the most reliable ways to steer Claude's format, tone, and structure. Instead of only describing what a good answer should look like, demonstrate the pattern you want Claude to reproduce. A useful mental model: Instruction → tells Claude the rule.

    Read
  2. 02
    Claude Prompting Study · 3 min

    Add Context

    Add context to improve performance. Claude can follow an instruction more intelligently when it understands the reason or motivation behind it. Context helps the model infer priorities and generalize beyond the exact wording of the instruction.

    Read
  3. 03
    Claude Blog Study · 3 min

    Steering Claude Code

    Put each Claude Code instruction in the control surface where it naturally belongs. The practical decision is not “how do I phrase this prompt?” It is: should this be always-on context, a scoped rule, a repeatable skill, an isolated subagent task, or a deterministic hook/permission?

    Read
  4. 04
    Claude Prompting Study · 3 min

    Be Clear and Direct

    Be clear and direct. Claude performs better when the prompt makes the desired result obvious. A strong prompt names the goal, adds the minimum context Claude needs, defines constraints, and specifies the output format.

    Read
Path 02

Design agent workflows

Move from one capable assistant to structured work with checkpoints, delegation, and explicit verification.

  1. 01
    Claude Blog Study · 3 min

    Parallelize the Work, Not the Uncertainty

    Key concept: Parallelize independent work, then verify before synthesis.

    Read
  2. 02
    Claude Blog Study · 3 min

    When to Use Subagents

    Key concept: Use subagents when context isolation, parallel execution, or a fresh perspective creates a real advantage.

    Read
  3. 03
    Claude Blog Study · 3 min

    When to Use Multi-Agent Systems

    Key concept: Use multiple agents only when one agent has a clear limitation.

    Read
  4. 04
    Claude Blog Study · 3 min

    Agent Workflow Patterns

    Pick the agent workflow pattern before you build the agent.

    Read
Path 03

Make implementation reliable

Explore the operating choices that turn an impressive demonstration into a workflow people can safely depend on.

  1. 01
    Claude Blog Study · 3 min

    Risk-Weighted Autonomy

    Risk-weighted autonomy: give agents more freedom for routine, reversible, bounded actions and require stronger human review when actions are external, destructive, financial, legally meaningful, production-facing, or difficult to reverse.

    Read
  2. 02
    Claude Blog Study · 3 min

    Put Policy in the Path

    Key concept: Move sensitivedata governance from written guidance into an enforced checkpoint on the data path before information reaches the AI model.

    Read
  3. 03
    Claude Blog Study · 3 min

    Trustworthy Self-Service Analytics

    Key concept: Reliable analytics agents need one governed path to the answer—not unrestricted access to every table and historical query.

    Read
  4. 04
    Claude Blog Study · 3 min

    Bounded Risk for Agentic AI

    Key concept: The goal of AI governance is not zero risk. It is to make agentic risk visible, bounded, and manageable.

    Read

Already know what you need?

Search the complete collection.

Use the search control in the header from any page, or open Studies for advanced filters and sorting.

Browse all studies