eric/weatherreporter
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 2 Wiki Activity

Harden PromptKit upgrade integration

Browse Source
This commit is contained in:
Eric Rakestraw
2026-08-25 23:46:21 +00:00
parent 7005688b80
commit 0bafbcb21f
12 changed files with 314 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
docs/cli.md
View File

@@ -106,7 +106,10 @@ period, prompt version, timezone, and status. Successful output has an absolute
When available, the summary also includes the effective `profileId`,
`backendId`, `modelName`, `sourceWarnings`, `validationStatus`, requested
`llmDebugPath`, and compact Distributor `notification` result. It does not
`repairAttempts`, `llmDebugPath`, and compact Distributor `notification`
result. `repairAttempts` is `0` when the initial output passed validation,
positive when PromptKit made corrective generation calls, and omitted when
validation did not complete. The summary does not
include historical or transient artifact paths such as metadata, prompt input,
raw generated text, render context, or notification receipts.
@@ -115,7 +118,8 @@ raw generated text, render context, or notification receipts.
A run summary contains `command`, `batch`, `status`, `startedAt`, `finishedAt`,
`total`, `succeeded`, `failed`, and a `reports` array. Each report item includes
its identity, status, effective profile and model details when available,
source warnings, validation status, and absolute `outputPath` after publication.
source warnings, validation status, repair-attempt count when validation
completed, and absolute `outputPath` after publication.
The top-level summary may also contain a batch `notification` object and
`error`. Batch status is `failed` if any report or the batch notification fails.
The `total`, `succeeded`, and `failed` counters describe report items only, so
@@ -143,7 +147,8 @@ A comparison summary contains these fields in this order: `command`,
successful `results[].reportPath` are absolute. `results` preserves the
supplied profile order and each item contains `position`, `profileId`, optional
`backendId`, `modelName`, `status`, optional `validationStatus`, optional
`reportPath`, optional `llmDebugPath`, and optional safe `error`.
`repairAttempts`, optional `reportPath`, optional `llmDebugPath`, and optional
safe `error`. The repair-attempt semantics match the generate summary.
The comparison status is `succeeded` only when every selected profile succeeds
and the bundle is published. Individual profile failures still publish a
@@ -157,7 +162,8 @@ published Promptkit category; destination failures use `destination_<kind>`;
and committed cleanup failures use `publication_cleanup` with a message that
states whether a complete prior bundle, partial remnants, or no prior bundle
remains, or that recovery state could not be inspected. It does not expose
provider diagnostics, filesystem causes, or recovery paths. See the
provider diagnostics, filesystem causes, or recovery paths. A provider HTTP
failure may include its numeric status in the safe message. See the
[comparison bundle contract](integrations/comparison-bundle.md) for durable
artifact fields and failure invariants.