01Text
Claude Sonnet 5.5
Faster model for clearly bounded knowledge and implementation work.
ArchiveWeek 40/2026 · 4 October 2026
5 AI toolsof the week
Text, design, data / knowledge, research and agents: one tool each, checked for value, cost and data protection.
The week in brief
This week AI becomes valuable where an answer turns into controllable execution. Claude Sonnet 5.5 speeds up clearly bounded knowledge work, while Strands Decider 2B carries out small routine decisions locally and with a confidence score. Suno Speech lowers the barrier for audio prototypes, Slashspace connects research in a local canvas and Dots shows the potential of persistent cloud agents. Data flows, permissions, regional availability and human approval remain decisive.
- Claude Sonnet 5.5 (8.8/10): strong for clearly bounded work, but only with a clean data and review process.
- Trend: from the single answer to controllable execution with local decisions, permissions and escalations.
- Rule of thumb: start with test data, minimal permissions, measurable quality criteria and a human approval point.
This issue dates from before the English edition launched in October 2026. Summaries and short descriptions are translated; the detailed reviews are available in German. Detailed review in German →
02Design
Suno Speech Beta
Spoken audio with an original music bed as one continuous track.
03Data / Knowledge
Strands Decider 2B
Open-source decision model for local triage, routing and guardrails.
04Research
Slashspace 3.3
Local-first canvas for multi-model research, sources and orchestration.
05Agents
Dots by OpenAI
Persistent cloud agents with app connections, rules and an activity view.
The detailed version with all research notes is available in German.