Merge pull request #110 from microsoft/fix/suppress-lowyield-agent-findings

review: stop reference-less agent findings in privacy and UI/accessibility leaves

microsoft/skills/review/al-privacy-review.md

@@ -68,7 +68,7 @@ Set `confidence` to:
 - `medium` when detection relies on heuristics or when any frontmatter dimension was `unknown`.
 - `low` when the finding is an advisory derived only from applicability.
 
-After evaluating each worklist entry, also consider whether the diff exhibits a privacy defect the agent recognises from its general AL knowledge that no knowledge file in the worklist covers. Such candidates are agent findings within this skill's domain — emit them with `references: []`, an `id` slug prefixed with `agent:`, `confidence` capped at `medium`, `severity` capped at `minor` (agent findings are advisory and non-gating), and a `message` that is self-contained (describing both the issue and a concrete recommendation, since there is no knowledge-file footer for the consumer to fall back on). Hold every candidate to the precision bar in `skills/do.md` (*Agent findings*): emit only a concrete, material privacy defect a knowledgeable BC reviewer would agree is wrong — steelman it first and drop anything stylistic, speculative, dependent on code outside the diff, or merely a valid alternative; when in doubt, omit. The scope is strictly privacy; defects outside this domain belong to other leaves and MUST NOT be emitted here. Before emitting, check the worklist for a knowledge file that matches the candidate — if one exists, upgrade the candidate to a knowledge-backed finding instead. See `skills/do.md` for the full contract.
+This leaf emits only knowledge-backed privacy findings. Do NOT emit reference-less `agent:` findings in this domain: online evaluation shows the privacy agent-finding channel yields almost no accepted findings and a high volume of dismissed noise, so a privacy concern that no worklist knowledge file covers is omitted here rather than emitted with `references: []`. When you spot a material privacy defect no article covers, the durable fix is to add a knowledge article in BCQuality (per the online-eval self-improvement loop) so this leaf can cite it — not a one-off reference-less finding. Before treating a candidate as uncovered, check the worklist for a knowledge file that matches it; if one exists, emit it as a knowledge-backed finding. See `skills/do.md` for the full contract.
 
 For every emitted finding, decide whether the fix is mechanical. A fix is mechanical when it is small, local, and unambiguous from the diff context (for example: delete unreachable lines; replace `Count() > 0` with `not IsEmpty()`; move a local `Label` to object scope; add a missing `ToolTip`, `OptionCaption`, or `DataClassification`; replace a string-concatenated `Error` with a Label-backed call; change an over-broad permission token; or add an obvious `else`/guard branch). For mechanical findings, emit `findings[].suggested-code` with the literal replacement for the source lines indicated by `location`. The payload must be a verbatim replacement — no diff markers, no fences, no commentary — that the consumer can render as a one-click suggestion. When a `.good.al` companion exists and the diff context matches the `.bad.al` shape, adapt the `.good.al` replacement into `suggested-code`.
 

microsoft/skills/review/al-ui-review.md

@@ -61,7 +61,7 @@ Set `confidence` to:
 - `medium` when detection relies on heuristics (judging whether a caption is a noun phrase or a sentence phrase) or when any frontmatter dimension was `unknown`.
 - `low` when the finding is an advisory derived only from applicability.
 
-After evaluating each worklist entry, also consider whether the diff exhibits a UI defect the agent recognises from its general AL knowledge that no knowledge file in the worklist covers. Such candidates are agent findings within this skill's domain — emit them with `references: []`, an `id` slug prefixed with `agent:`, `confidence` capped at `medium`, `severity` capped at `minor` (agent findings are advisory and non-gating), and a `message` that is self-contained (describing both the issue and a concrete recommendation, since there is no knowledge-file footer for the consumer to fall back on). Hold every candidate to the precision bar in `skills/do.md` (*Agent findings*): emit only a concrete, material UI defect a knowledgeable BC reviewer would agree is wrong — steelman it first and drop anything speculative, dependent on code outside the diff, or merely a valid alternative; when in doubt, omit. The scope is strictly UI; defects outside this domain belong to other leaves and MUST NOT be emitted here. Before emitting, check the worklist for a knowledge file that matches the candidate — if one exists, upgrade the candidate to a knowledge-backed finding instead. See `skills/do.md` for the full contract.
+This leaf emits only knowledge-backed UI and accessibility findings. Do NOT emit reference-less `agent:` findings in this domain: online evaluation shows the UI/accessibility agent-finding channel yields almost no accepted findings and a high volume of dismissed noise, so a UI or accessibility concern that no worklist knowledge file covers is omitted here rather than emitted with `references: []`. When you spot a material UI or accessibility defect no article covers, the durable fix is to add a knowledge article in BCQuality (per the online-eval self-improvement loop) so this leaf can cite it — not a one-off reference-less finding. Before treating a candidate as uncovered, check the worklist for a knowledge file that matches it; if one exists, emit it as a knowledge-backed finding. See `skills/do.md` for the full contract.
 
 For every emitted finding, decide whether the fix is mechanical. A fix is mechanical when it is small, local, and unambiguous from the diff context (for example: delete unreachable lines; replace `Count() > 0` with `not IsEmpty()`; move a local `Label` to object scope; add a missing `ToolTip`, `OptionCaption`, or `DataClassification`; replace a string-concatenated `Error` with a Label-backed call; change an over-broad permission token; or add an obvious `else`/guard branch). For mechanical findings, emit `findings[].suggested-code` with the literal replacement for the source lines indicated by `location`. The payload must be a verbatim replacement — no diff markers, no fences, no commentary — that the consumer can render as a one-click suggestion. When a `.good.al` companion exists and the diff context matches the `.bad.al` shape, adapt the `.good.al` replacement into `suggested-code`.