docs(ci): reduce sequence-diagram audit churn (anti-reword + canonical terms)#922
Merged
Merged
Conversation
…agram audit prompt The weekly verify-sequence-diagrams audit re-flagged diagrams that were already correct, because a second AI pass models the same code with different-but-equally-valid wording/granularity (e.g. Class vs Type, Assembly vs SIA/EIA participants, swapped valid examples, A2A vs SSE labels). Add a 'What NOT to change' section forbidding rewording, re-modeling at a different granularity, swapping still-valid examples, and adding deliberately-omitted detail unless the code actually contradicts the diagram. Add a canonical-terminology section pinning 'DeliveryMode.Stream SSE' and forbidding the 'A2A' label so generation and audit passes agree. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 56262116-a37d-4365-bb56-f489b6cd5e2e
tracyboehrer
requested review from
a team,
ceciliaavila and
sw-joelmut
as code owners
July 16, 2026 14:08
Contributor
There was a problem hiding this comment.
Pull request overview
This PR updates the CI prompt used by the weekly verify-sequence-diagrams audit to reduce “interpretation-only” churn by explicitly prohibiting rewording/re-modeling of already-correct Mermaid diagrams, and by pinning canonical terminology so future audit passes converge on consistent labels.
Changes:
- Adds a “What NOT to change (avoid audit churn)” section that gates edits on code-contradiction evidence rather than alternative-but-valid phrasing/modeling.
- Adds a “Canonical terminology” section intended to standardize the SSE streaming wording and avoid reintroducing “A2A” labeling.
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
tracyboehrer
enabled auto-merge
July 16, 2026 14:16
rodrigobr-msft
approved these changes
Jul 16, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why
The weekly
verify-sequence-diagramsaudit keeps re-flagging Mermaid diagrams that are already correct. Because the diagrams were originally authored by an AI pass, a second AI audit pass finds different-but-equally-valid ways to model the same code and "corrects" them — churn, not fixes (see the interpretation-only edits in #921:Class→Type,Assemblysplit intoSIA/EIA, swapped-but-valid converter examples,A2A↔SSElabels).Change
Doc/CI-only edit to
.github/prompts/verify-sequence-diagrams.prompt.md:What NOT to change (avoid audit churn)— forbids rewording accurate labels, re-modeling participant granularity, swapping still-valid examples, and adding deliberately-omitted detail unless the code contradicts the diagram. Gated by: "Can I point to a line the diagram states is true but isn't?"Canonical terminology— pinsDeliveryMode.StreamSSE and forbids reintroducing theA2Alabel, so generation and audit passes agree on wording.Effect
UserSignInFailureHandler, renamed calls, moved paths).