Also called: runtime block, runtime_block_json
A Voice DNA runtime block is a compact, machine-readable rendering of a writing profile, formatted to be loaded directly into a language model's system prompt so an agent can adopt a person's voice without a human pasting instructions.
It is the profile in the shape a system prompt can consume, rather than the shape a human reads.
A writing profile has two audiences. A human reads the long-form report to understand their own patterns. A model needs the same information compressed, unambiguous and positioned where system instructions go. The runtime block is the second rendering.
The practical consequence is agent-native delivery. Because the block is data rather than prose, an agent can fetch it over an API or MCP tool call and install it, so the person's voice follows them into automations they never open a chat window for.
| Rendering | Audience | Shape | Where it is used |
|---|---|---|---|
| Writing Style Profile report | The author | Long-form prose with examples | Reading, reviewing, editing your own patterns |
| Runtime block | A language model | Compact structured block, mode-sliced | System prompt, agent config, MCP tool response |
Cite this page
MyWritingTwin. “Voice DNA runtime block.” Last updated 2026-09-15. https://www.mywritingtwin.com/glossary/voice-dna-runtime-block