Anthropic API

Entwicklungsaufgaben, Projektkontext und überprüfbare Ergebnisse mit Anthropic API.

Anthropic API

Agent SDK, Managed Agents, or the raw API Englisch Anthropic API
Four ways to build on Claude, separated by one question: who runs the agent loop and the sandbox. Answer that and the choice mostly makes itself.
Batching and cost control Englisch Anthropic API
Half price for work that can wait an hour. The catch is a hard 24-hour expiry, and it is the only cost lever that needs no prompt changes.
Building an MCP server: exposing your thing to the model Englisch Anthropic API
A server is a small adapter between the model and something you own. The hard parts are not the protocol; they are the trust boundary you just opened.
Choosing a model Englisch Anthropic API
Start at the default, move only for a reason, and match the tier to the step rather than to the project.
Context editing, compaction, or memory Englisch Anthropic API
Three mechanisms for a conversation that outgrows its window. One clears, one summarises, one stores — and they compose rather than compete.
Do you actually need an agent? Englisch Anthropic API
Anthropic's own advice is to find the simplest thing that works, which "might mean not building agentic systems at all". Usually one call is enough.
Error handling that distinguishes retryable from not Englisch Anthropic API
A 529 deserves a retry, a 400 never earns one. Sorting the API's errors into those two bins is most of production error handling.
Evaluating an agent: what "working" means Englisch Anthropic API
Anthropic's own advice is volume over polish: more cases graded roughly beats few graded well. For an agent, grade the trajectory and not only the answer.
Migrating to a newer model Englisch Anthropic API
A model swap is not a string change. Parameters break loudly, behaviour changes quietly, and the quiet half is where migrations actually fail.
Prompt caching that actually hits Englisch Anthropic API
Reads cost a tenth of normal input, so caching pays from the second request. A prompt under the minimum is silently not cached, with no error.
Streaming and long turns: the 200 is not the finish line Englisch Anthropic API
Streaming trades one big wait for many small events, and moves failure to after the response starts. The event flow and the mid-stream error are the two things to get right.
Structured output Englisch Anthropic API
Guaranteed-valid JSON is not guaranteed-valid data. Half your schema is stripped before the model ever sees it, and your code enforces the rest.
Writing a tool description a model can use Englisch Anthropic API
Anthropic calls the description "by far the most important factor in tool performance". Most are one line long. Aim for three to four sentences.
Tool use without the common foot-guns Englisch Anthropic API
Ask for the weather with no city and it may invent one, plus a unit you never asked about. Four failures that are cheap to prevent and expensive to debug.
Thinking, reasoning and "effort" — what these controls do Englisch Anthropic API
Thinking is billed output you mostly never read. The control moved from a token budget to an effort level, and on current models the old one errors.
Compare API cost and context correctly Englisch Anthropic API + OpenAI API + Gemini API
Separate cumulative usage from peak context, account for the chosen pricing conditions, and measure quality independently.
Workflow: debugging a failure Englisch Claude + Claude Code + Anthropic API
Narrowing costs four times what diagnosing costs. Put the cheap model on the search and the expensive one on the thirty seconds that decide the answer.
Workflow: a refactor across several files Englisch Claude + Claude Code + Anthropic API
The plan is 1% of the tokens and decides the other 99%. Everything else is mechanical, and mechanical work does not need the expensive model.
Workflow: researching and synthesising Englisch Claude + Claude Code + Anthropic API
Reading thirty sources costs eight times what thinking about them costs. This is the workflow where model choice saves the most.
Workflow: reviewing a change Englisch Claude + Claude Code + Anthropic API
Three steps, three different models. Reading is not judging, and paying Opus rates to read a diff is the commonest waste in this workflow.

Modelle, Zugang und Limits prüfen