OpenVoiceOS / OpenVoiceOS/architecture
Add a `.prompt` resource role for language-model prompts (OVOS-INTENT-2 V3)
Nobody has claimed this yet.
- Dominant language
- No language data
- Stars
- 3
- Forks
- 0
- Avg merge
- 15h 20m
- Merged PRs (30d)
- 90
Description
Tracking issue for PR #4 — OVOS-INTENT-2 v3: the .prompt resource role
Local and hosted language models are now ubiquitous. The prompt a skill sends to that model is a first-class skill resource and, like every other resource, should be localizable: shipped per language, overridable, resolved through the same precedence chain. The current OVOS-INTENT-2 resource roles have no place for it.
Proposal
Add a sixth resource role to OVOS-INTENT-2 — .prompt — taking the spec to version 3.
A .prompt lives under locale/<lang>/ like any other resource and differs from .dialog in three deliberate ways:
- Whole-file — the entire file content, verbatim, is one prompt.
- Plain text, not a template — no OVOS-INTENT-1 grammar. The only special construct is
{name}substitution. - Conservative slots —
{name}is substituted only when well-formed, supplied, and not inside a fenced code block (CommonMark-aligned). HTML-style<!-- … -->comments are stripped before the prompt reaches the model. Unfilled slots remain as literal text.
V3 is additive over V2.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reviewing PR #4 and the existing OVOS-INTENT-2 V2 resource-role specification. Confirm how the proposed .prompt role, localization, precedence, substitution rules, fenced-code handling, and comment stripping should be represented; done means the architecture specification is updated to a consistent V3.
Written by the indexing model from the issue text.
Assessment
- Domain
- documentation
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 25/100