Verbosity Steering
# Verbosity steering — L0 through L4
These are the system-prompt fragments that implement Headroom's output-token
shaper. Append the selected level to the **tail** of the system prompt
(after any `cache_control` breakpoint, so the cached prefix stays warm).
The default level is **L2**. The five levels are cumulative: L2 = L1 + its
own rules; L3 = L2 + its own rules; etc.
> **Why the tail, not the head.** Prompt caching is a prefix match. Any byte
> change ahead of a `cache_control` breakpoint invalidates everything after
> it. Prepending steering text would bust the provider prefix cache and cost
> more than it saves. Appending after the last system block leaves the cached
> prefix byte-identical; only the small, byte-stable steering block is
> reprocessed.
---
## L0 — off
(No text appended. Write naturally.)
---
## L1 — no ceremonywhen to use it
Community prompt sourced from the open-source GitHub repo roman-ryzenadvanced/headroom-skill (Apache-2.0). A "Verbosity Steering" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.
tags
roleplaycommunitygeneral
source
roman-ryzenadvanced/headroom-skill · Apache-2.0