eric/promptkit
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Escape schema resources and reuse compiled plans

Browse Source
This commit is contained in:
Eric Rakestraw
2026-08-11 23:05:11 +00:00
parent a93b799236
commit 20d3e3b5ee
11 changed files with 458 additions and 270 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
docs/internal/runner.md
View File

@@ -20,8 +20,9 @@ and override semantics consumed by the runner.
profiles, backend resolution, artifacts, rendering, model generation, and
validation. The root engine supplies one immutable registry containing the
built-in backend and validated consumer additions, one engine-local run
admitter, and a model client wrapped by the same capacity manager. Schema
documents are loaded through the validator's optional schema-loader interface.
admitter, and a model client wrapped by the same capacity manager. Validation
plans and provider-facing schema metadata come from the validator's preparation
interface.
An output repairer can be injected internally, but the ordinary runner
constructor does not enable one.
@@ -77,7 +78,8 @@ performs only the work needed to validate routing and admission:
The completion phase consumes that state without reloading the prompt,
profile, or backend:
1. load structured-output schema metadata when required;
1. create one operation-local validation plan and derive structured-output
schema metadata from it when required;
2. load and hash input artifacts;
3. render messages and the prompt-defined session;
4. apply any direct session ID;
@@ -88,7 +90,10 @@ profile, or backend:
`Run` performs backend admission between the phases. This structure preserves
one execution-precedence and error-ordering implementation while allowing a
full backend pool to reject work before expensive schema, artifact, and
rendering operations.
rendering operations. `Prepare` discards the plan after returning its public
metadata. `Run` retains the plan through initial and repaired-output validation
and discards it when the operation ends. Prepared execution stores the same
kind of plan only in its private payload.
Pointer-based numeric overrides preserve an explicit zero. Invalid negative or
out-of-range values fail as invalid requests. Endpoint overrides do not change
@@ -119,8 +124,9 @@ its `RunAdmitter` to reserve capacity for the effective backend ID. A nil
admitter is an internal unlimited fallback. After successful admission, `Run`
immediately defers the returned release function, performs the completion
phase, makes one initial generation call, builds the named output artifact,
and validates that artifact. Invalid generated content remains a validation
result; an inability to perform validation is an operational error.
and validates that artifact with the plan compiled during completion. Invalid
generated content remains a validation result; an inability to perform
validation is an operational error.
The admission lease covers completion-phase preparation, initial generation,
validation, every repair, and every exit. It bounds accepted work without
@@ -134,8 +140,8 @@ target and session ID, validation errors, prior output, and structured-output
specification. The default repairer uses the same wrapped client as initial
generation, so each repair reacquires the selected backend's active permit
while remaining inside its original admission lease. Repair never performs a
second bounded admission. This capability remains internal and is not a public
option.
second bounded admission, and repaired outputs use the operation's existing
validation plan. This capability remains internal and is not a public option.
A successful result includes the output artifact and raw output, validation
state, effective session ID, prompt and rendered-prompt hashes, selected