4.0 KiB
Audita Prompt Registry
internal/prompts embeds seven PromptKit YAML definitions under
assets/prompts. The IDs and selected version remain:
modules.glossary.proposal(v1)modules.homophones.proposal(v1)modules.spoken_word.proposal(v1)modules.grammar.proposal(v1)validators.spoken_form_plausibility(v1)validators.meaning_reversal_review(v1)validators.editorial_review(v1)
PromptKit reads the embedded definitions as shipped defaults and renders their
messages from named string variables. Operator documents in profiles_dir may
contribute native keyed prompt definitions and override an embedded
prompt/version, including its messages and definition-level output settings.
The effective operator text is captured when the runtime is constructed; the
embedded assets remain the source of Audita's shipped task instructions. The
role-bound adapter supplies
HardeningPolicy alongside ProposalContract and ValidationContract from
embedded fragments under assets/shared, after caller variables so callers
cannot replace those instructions. Common literal-span and identity guidance
lives in the proposal fragment; shared candidate-reference, context, and
confidence guidance lives in the validation fragment. Module-specific policies
and examples remain in YAML. Proposal prompts receive the section JSON, glossary
JSON where used, and transcript description block. Validator prompts receive
the correction payload and description block.
The embedded prompt assets retain the shipped task instructions and
output-schema association. An operator override controls which template
variables its messages reference: supplying a shared variable does not append
its text when the effective message omits that reference. The role-bound runtime
supplies the complete JSON Schema output contract and repair count; the
definition does not own an operator-facing repair default.
The correction-set schema requires a string replacement and accepts an empty
string. Proposal guidance uses it for eligible span deletion and permits literal
spacing edits; validator guidance evaluates the full before/after segment under
each validator's existing remit. A blank resulting segment remains a domain
rejection, not a structural schema error.
internal/framework/responseschema remains the only embedded source for
correction_set.v1.json and validator_decision_set.v1.json. It exposes that
filesystem to PromptKit and retains Audita's schema-key lookup. PromptKit's
prepared execution supplies definition and rendered hashes; Audita does not
recompute or promise equality with hashes from the removed template renderer.
The small Audita registry records only prompt ID, version, source, and embedded
path. Prepared runtime metadata belongs in diagnostics rather than the registry.
The structured-call contract carries prompt ID/version, string variables, response-schema key, and logical-call attribution. Module and validator callers pass these values through the role-bound adapter. See the LLM runtime for runtime assembly and configuration for operator-selected profiles.
Proposal instructions describe literal matching, each module's replacement policy,
source-based proposal identity, and confidence in whether an edit is warranted.
Proposal id and validator correction_index values are short, one-based
references scoped to each request, not internal transcript/proposal identities.
Mapping belongs to proposal generation and
validator reconciliation. Validator instructions distinguish
confidence in the verdict from the verdict itself. Compact output examples
illustrate the supplied schema without requiring fixed JSON formatting or key
order. The grammar article exception and glossary mishearing exception apply
consistently across their instructions.
Only the three active validator definitions are retained. The unused internal
grammar_review and spoken_word_review definitions and type selections were
removed; built-in chains and operator configuration are unchanged.