Plan correction repair after validation

This commit is contained in:
2026-11-11 00:05:04 +00:00
parent 8946a82407
commit d91bb658c9
2 changed files with 386 additions and 10 deletions

View File

@@ -1,6 +1,6 @@
# Deferred Feature Work
**Last updated:** 2026-10-11
**Last updated:** 2026-11-10
**Status:** Deferred or exploratory work; not implemented or scheduled. These
entries preserve work beyond completed feature and library upgrades.
@@ -10,10 +10,11 @@ implementing every capability.
## Domain-Aware Response Recovery
The implemented [partial-success decision reconciliation](../internal/validators.md#llm-validator-batching)
preserves usable decisions without another model call. Guided recovery remains
deferred pending evidence from failures that remain. General per-entry
salvage from structurally invalid responses is also a separate design question;
the accepted reconciliation feature retains whole-response structural validation.
preserves usable decisions without another model call. Revision of rejected
correction proposals is now an agreed feature direction in the
[correction-repair roadmap](correction-repair.md), with its own open decisions.
That work includes substantive validator feedback and explicit proposal withdrawal.
It is distinct from recovering an unavailable validator decision.
Consider bounded feedback and regeneration for model responses that satisfy the
structural schema but leave requested decisions unavailable after reconciliation.
@@ -30,11 +31,15 @@ secrets, and how regeneration fits bounded budgets and operation deadlines. Reus
shared library capabilities where appropriate; do not introduce an uncoordinated
outer retry loop that multiplies existing retry/repair budgets.
Ordinary editorial rejection is a valid decision, not a reason to regenerate until
an edit is approved. Recovery must preserve required validation, candidate identity,
partial-success safety, and truthful error/recovery reporting. Validate with
controlled responses that distinguish recoverable protocol defects from valid
negative decisions. Exact APIs, budgets, and sequencing remain undecided.
Editorial rejection remains a valid decision about the evaluated proposal. The
correction-repair roadmap owns any subsequent revision of that proposal and full
revalidation; this deferred entry concerns unavailable decisions. Decision recovery
must preserve required validation, candidate identity, partial-success safety,
and truthful error/recovery reporting. Validate with controlled responses that
distinguish protocol defects from valid negative decisions. Exact APIs, budgets,
and sequencing remain undecided. General per-entry salvage from structurally
invalid responses also remains a separate design question; current reconciliation
retains whole-response structural validation.
## Independent Whole-Pipeline Traversal by Chunk