Refresh native workflow orientation and planning status

This commit is contained in:
2026-11-22 06:57:27 +00:00
parent b76d511571
commit 633ebd282f
7 changed files with 165 additions and 111 deletions

View File

@@ -7,12 +7,18 @@ PromptKit/ConfigKit/CommandKit adoption, continuation, admission, cache, output,
and debug decisions. Items below are ordered roughly by value and specificity,
not as committed release dates or prerequisites for that migration.
Native shared loading/commands, complete assessment/continuation, concurrent
ready lane completion, accepted-result reuse, and private capture are implemented;
see the [implementation evidence](implementation.md) for final acceptance status.
They are not deferred platform work. Human model-quality evaluation, broader
reference/context policies, and storage/distribution extensions below remain separate.
## Near-Term D&D Pipeline
### Evaluate Spell Extraction And Normalization
- Evaluate extraction, normalization, and correction quality against a
human-reviewed transcript set before and after the native migration.
human-reviewed transcript set, comparing retained baselines with the native path.
- Maintain a small set of human-reviewed transcripts and outputs for prompt,
validator, and normalizer development. Treat model-quality review as an
iterative human evaluation aid, not a deterministic correctness gate.
@@ -78,9 +84,9 @@ The sections below keep broader extensions deferred.
benefits from model judgment and deterministic checks are insufficient.
- Consider additional validator diagnostic views only if a concrete need remains
after the selected [native observations and debug capture](promptkit.md#inspection-and-execution-bookkeeping).
- Add validator compatibility metadata if deployments need config-time proof
that a validator is suitable for a particular stage, module, or artifact
kind.
- Existing typed resolution already checks validator target, artifact kind,
and exact Go-type compatibility. Add further semantic suitability metadata
only if a deployment needs a contract beyond those implemented checks.
- Add media-type validators when non-JSON artifact representations are
introduced.
@@ -88,8 +94,9 @@ The sections below keep broader extensions deferred.
- Make prior-run artifacts easier to bind as references without changing the
existing module-facing reference-item contract.
- Add structured or parsed references, such as typed NPC registries, rosters,
or spell catalogs, when opaque UTF-8 prompt material is no longer sufficient.
- Existing registry/catalog consumers already parse and validate typed contextual
projections. Consider a framework-level parsed-reference facility only when
concrete consumers need shared behavior beyond those module-owned resolvers.
- Add per-slot or per-chunk inclusion policies so large references are not
repeated in every prompt unnecessarily.
- Add token budgeting and model context-window management for reference
@@ -146,7 +153,8 @@ section only after a concrete workflow, contract, and priority emerge.
- Additional input adapters, such as Markdown or note-export formats.
- Additional output encoders.
- Concurrent cross-lane entity normalization or broader workflow composition.
- Broader workflow composition only when a concrete need justifies revisiting
the fixed-shape pipeline decision; ready cross-lane completion already runs concurrently.
- Batching or specialized context-window controls for LLM-backed validators.
### Distribution And Operations