34950f875e
Merge pull request #1094 from dotnet/automated/gh-aw-upgrade
...
chore: upgrade gh-aw runtime
2026-08-31 15:17:37 -07:00
ef5f7cada6
Merge pull request #1095 from dotnet/bot/weekly-version-sync
...
Weekly plugin version sync
2026-08-31 15:17:17 -07:00
f775756a79
Improve cross-model test migration outcomes ( #1096 )
...
* Improve test migration skill outcomes
Address cross-model routing and execution failures found in retained evaluation evidence, and align the xUnit companion-package rubric with feed-compatible versions.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Cover MTP method filter routing
Include the method-filter token in the VSTest-to-MTP skill description so method-specific migration requests activate consistently.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Clarify MSTest SDK v4 discovery guidance
Scope the missing Test SDK reference to MSTest.Sdk v4 and remove a fixed package-version example in favor of feed-resolved and CPM-aware guidance.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Clarify xUnit MTP runner preservation
State that the migration preserves the MTP runner for projects currently using the YTest shim, rather than preserving the shim package itself.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Avoid xUnit package major assumptions
Resolve companion compatibility from package constraints and configured feeds instead of examples that conflate the xUnit product generation with NuGet major versions.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* List all MSTest SDK discovery paths
Align the CI response contract with the three documented choices for transitional VSTest discovery and MTP-native execution.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Fix xUnit v3 evaluation evidence
Replace a contradictory async-void output grader with a file-state check, strengthen deterministic BeforeAfter signature checks, and require explicit final evidence for both migration paths.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Align MSTest SDK discovery guidance
Correct the behavioral symptom table to describe the v4-only removal and all three documented migration paths consistently.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Make BeforeAfter eval name agnostic
Validate the required xUnit v3 override and base-call shapes without hardcoding local parameter names.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Strengthen Luna migration outcomes
Add complete advisory output contracts, correct focused-fix execution decisions, require concrete MTP package and filter guidance, and make xUnit package and source evidence explicit.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Relax BeforeAfter source formatting
Keep deterministic IXunitTest source validation while relying on compilation and test execution for qualified and multiline signature correctness.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-31 17:15:52 +00:00
a7a6170880
Improve template and JSON skill guidance ( #1097 )
...
* Improve template and JSON skill guidance
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Align template validation guidance
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Refine skills from evaluation evidence
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Clarify SDK and parse guidance
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Make local SDK setup self-contained
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Improve template eval discrimination
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Fix template comparison wording
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Harden multiline command graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Align multiline positive graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Collapse dry-run command graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Keep parse errors out of findings tables
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Stage template fixtures explicitly
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Enforce safe template command boundaries
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Clarify template pack target framework
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Cover inline negative command checks
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Align template framework selection guidance
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Scope workspace framework grader
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Require options in template commands
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-31 16:58:04 +00:00
250089d2dc
Cover xUnit-to-MSTest Parallelize eval pattern ( #1098 )
...
* Cover xUnit migration parallelization eval
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5b0aa9a5-aa8f-445f-8db0-0cc6795d03ac
* Improve xUnit migration consistency
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5b0aa9a5-aa8f-445f-8db0-0cc6795d03ac
* Preserve xUnit migration semantics
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5b0aa9a5-aa8f-445f-8db0-0cc6795d03ac
* Make migration evidence explicit
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5b0aa9a5-aa8f-445f-8db0-0cc6795d03ac
---------
Copilot-Session: 5b0aa9a5-aa8f-445f-8db0-0cc6795d03ac
2026-08-31 16:10:12 +00:00
6cc9e1e1fc
Weekly plugin version sync
2026-08-31 09:15:51 +00:00
d9d546b663
chore: upgrade gh-aw runtime
2026-08-31 09:12:53 +00:00
d68dd70857
Merge pull request #1084 from dotnet/abhitejjohn-surface-skill-value
...
Surface the Skill Value dashboard
2026-08-28 09:51:57 -07:00
1dd5e2b7d2
Improve test gap correctness and cost ( #1080 )
...
* Improve test gap correctness and cost
Require complete public-outcome inventories, suppress inert and unobservable mutation candidates, and bound focused execution and output.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Tighten test gap analysis execution scope
Reserve mutation execution for explicit verification, require complete authorization outcome ledgers, and calibrate exact-value and suite-strength reporting from the latest Luna evidence.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Suppress speculative test gap findings
Keep focused analyses within the named risk, require concrete edits to existing code, and cover complete money-math composition without future-only or trivial state findings.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Clarify observable error semantics
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Resolve remaining review wording
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Refine test gap scope and boundaries
Convert named risks into observable outcome allowlists, enumerate ordered guard partitions, and align the advisory rubric with static source-to-assertion evidence.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Align advisory evaluation and verification
Exclude derived non-monetary predicates, require canonical verification scripts for test additions, and remove the advisory rubric's incentive for unnecessary mutation execution.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Address test gap review clarity
Keep the output contract limited to response components and use clearer mutation terminology in the advisory rubric.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Correct test gap evidence and calibration
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Cover retry classifier edge partitions
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Reject already-killed mutation gaps
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Clarify retry cutoff rubric
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Keep test tagging prompts dormant
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
* Require production-change routing intent
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 3a626035-c70f-4ba6-97f0-e076b8a5fd1e
2026-08-28 09:48:58 -07:00
4617228a43
Merge pull request #1083 from dotnet/abhitejjohn-manual-dashboard-publish
...
Allow manual main dispatches to publish dashboard eval data
2026-08-28 09:32:59 -07:00
8d14fa2568
Merge pull request #1046 from dotnet/harden-min-threads-eval-graders
...
Pin min-threads and env-var ground truth in migrate-dotnet8-to-dotnet9 rubric
2026-08-28 09:15:47 -07:00
2b9056bd91
Separate dormancy evidence and improve affected evals ( #1079 )
...
* Separate dormancy from preference evidence
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Preserve legacy dashboard preference semantics
Default scenarios without schema v4 preference eligibility metadata to eligible so historical schema v3 records remain consistent with their original gate evidence.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 808c8105-e596-42c6-9dca-4c62cfcd490f
* Cover activation contract reporting precedence
Exercise the combined activation-contract failure and underpowered preference state so consolidation keeps the routing failure as the headline result.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 808c8105-e596-42c6-9dca-4c62cfcd490f
* Tighten dormancy scalar parsing
Require complete YAML false tokens so prefix values cannot alter preference eligibility, and keep activation-contract summary grammar correct.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 808c8105-e596-42c6-9dca-4c62cfcd490f
* Preserve legacy dashboard gate labels
Mark only schema v4 dashboard evidence with dormancy exclusions so historical schema v3 records retain their original all-stimulus vote labeling.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 808c8105-e596-42c6-9dca-4c62cfcd490f
* Improve skills from evaluation evidence
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 808c8105-e596-42c6-9dca-4c62cfcd490f
* Harden dormancy evaluation contracts
Match PyYAML dormancy parsing, preserve unmatched annotations, prioritize gate-eligible report evidence, and make the MTP Boolean grader exact without case sensitivity.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 808c8105-e596-42c6-9dca-4c62cfcd490f
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 808c8105-e596-42c6-9dca-4c62cfcd490f
2026-08-28 05:39:59 -07:00
6d96a6ec1b
Document Skill Value dashboard
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: e23e5efe-713f-4898-a158-8970f70fdcaf
2026-08-27 13:12:00 -07:00
be7b56028c
Fix Codex discovery of the dotnet-msbuild binlog MCP server ( #1070 )
...
* Fix Codex discovery of the dotnet-msbuild binlog MCP server
.codex-plugin/plugin.json declared "mcpServers": "./.mcp.json", but the file
was packaged at .codex-plugin/.mcp.json. Codex resolves manifest resource
paths against the plugin root, so it looked for
plugins/dotnet-msbuild/.mcp.json and never found the binlog server.
Declare the server inline in .codex-plugin/plugin.json, matching the root
plugin.json and .claude-plugin/plugin.json, and drop the unreachable file.
Add a packaging regression check to skill-validator: every companion manifest
must declare the same MCP servers as the root plugin.json, and a manifest
referencing an external .mcp.json must resolve it from the plugin root the way
hosts do. skill-check.yml already runs `skill-validator check` over plugins/*
on every PR, so this now blocks in CI. A test also loads the shipped
dotnet-msbuild manifests and asserts binlog is present in each.
Fixes #1069
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com >
* Report non-object JSON roots as validation errors
JsonElement.TryGetProperty throws InvalidOperationException when the root
value is not an object, so a manifest or referenced .mcp.json that is valid
JSON but not an object (null, array, string) crashed skill-validator instead
of producing a validation error.
Check the root kind while reading and surface it as a structured error.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com >
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com >
2026-08-27 17:58:26 +00:00
576639d85c
Preserve unflagged manual dashboard deploys
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 7f5635cf-ce89-4be6-b336-01512e957e9c
2026-08-27 10:21:16 -07:00
f8b2dca660
Correct ThreadPool env-var rubric
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-27 10:01:58 -07:00
91cf907b5e
Allow manual dashboard eval publishing
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-27 09:56:18 -07:00
7e5b5989f6
Clarify platform detection and MTP hot reload boundaries ( #1076 )
...
* Clarify platform and hot reload boundaries
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Address platform and hot reload review feedback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Tighten platform detection response scope
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Address follow-up platform review feedback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Improve platform and hot reload eval discrimination
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Allow multiline platform causality evidence
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Correct dotnet watch restart syntax
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Harden xUnit hot reload fixture
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Clarify watch project ownership
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Expand platform import-order coverage
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
* Complete platform evidence explanations
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d3d8e36b-fdc3-470f-81c8-dc8d696590fb
2026-08-27 15:59:34 +00:00
f6f2751abd
Clarify evaluation progress PR label ( #1082 )
...
* Clarify evaluation progress label
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d35e7696-f69c-4898-ad3e-c75fb02b993d
* Distinguish queued and running evaluations
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: d35e7696-f69c-4898-ad3e-c75fb02b993d
---------
Copilot-Session: d35e7696-f69c-4898-ad3e-c75fb02b993d
2026-08-27 14:41:18 +00:00
26caf0ecab
Scale coverage analysis to request scope ( #1078 )
...
* Scale coverage analysis to request scope
Gate CRAP and full reports behind explicit intent, reconcile coverage arithmetic, and strengthen only evidence-backed eval outcomes.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Address coverage eval review feedback
Broaden no-command guards for Markdown formatting and clarify plateau recommendations across both uncovered members.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: dc01c723-1c85-446e-873a-c4bd26fdb0e8
* Improve coverage analysis recovery
Use the latest cross-model evidence to recover from failed artifact reads, require source-safe branch interpretation, make target combinations concrete, and enforce CRAP-first refactoring risk ranking.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: dc01c723-1c85-446e-873a-c4bd26fdb0e8
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: dc01c723-1c85-446e-873a-c4bd26fdb0e8
2026-08-27 14:04:37 +00:00
94ca0ca748
Improve evaluation freshness and scheduled reliability ( #1081 )
...
* Improve evaluation freshness and reliability
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Address dashboard freshness review
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Preserve evidence commit fallback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5283cdb3-86e3-41d8-95a9-ebf6b7e0ccbc
* Keep watchdog regression assertion current
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5283cdb3-86e3-41d8-95a9-ebf6b7e0ccbc
* Add headroom for MSTest migration evaluation
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5283cdb3-86e3-41d8-95a9-ebf6b7e0ccbc
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5283cdb3-86e3-41d8-95a9-ebf6b7e0ccbc
2026-08-27 13:34:24 +00:00
f65b81bd71
Stabilize skill menu and evaluation retries ( #1072 )
...
* Stabilize skill evaluation retries
Reduce the dotnet-test skill menu below the runtime budget and retry only transient baseline/skilled session.idle timeouts without replacing successful evidence.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Harden evaluation retry matching
Normalize eval paths before retry matching and require stable shardKey identities so recovery fails closed when record identity is uncertain.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 4c7484eb-1fa5-4332-841e-e924031c75a2
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 4c7484eb-1fa5-4332-841e-e924031c75a2
2026-08-27 14:45:02 +02:00
47dd914532
Add Claude plugin manifests ( #1071 )
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-27 12:16:20 +00:00
0be2ef95a5
Fix nested testability override restoration ( #1077 )
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-27 12:14:04 +00:00
0c27460786
Clarify injected dependency scan exclusions ( #1074 )
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-27 12:11:42 +00:00
2c3d9807b0
Merge pull request #1059 from dotnet/abhitejjohn-skill-value-dashboard
...
Add per-skill "Skill Value" view to the evaluation dashboard
2026-08-27 04:35:15 -07:00
a77e7da573
Fix MSTest version and retry guidance ( #1075 )
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-27 11:31:27 +00:00
bce81f3cbd
Clarify missing pass data tooltip
...
Describe the observed absence of counted pass data without assuming that a skill has no boolean graders.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5b4ee7ec-7244-4668-a151-5abee0ef2c4a
2026-08-27 03:05:34 -07:00
09e5babf49
Reduce dotnet-test skill menu size
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 032f6f50-dd65-4805-b7af-78755f8bbc77
2026-08-27 11:58:26 +02:00
d3921f7418
Strengthen testability skill evaluations ( #1057 )
...
* Strengthen testability skill evaluations
Raise four dotnet-test evals to eight independent stimuli, add validated fixtures, and resolve code-testing-agent orphan fixtures without speculative routing changes.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Relax promo-code eval grader
Accept deterministic suffix values beyond one hard-coded literal and match common PascalCase test names.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 84f7c88c-c8e7-4d8f-96c9-421de725ab31
* Improve testability skill reliability
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 84f7c88c-c8e7-4d8f-96c9-421de725ab31
* Refine testability obstacle graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 84f7c88c-c8e7-4d8f-96c9-421de725ab31
* Accept qualified Random seams
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 84f7c88c-c8e7-4d8f-96c9-421de725ab31
* Improve ambient seam compatibility
Replace the C# 12 primary constructor in the copyable Scope sample with a conventional constructor so the guidance works in projects using older language versions.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 84f7c88c-c8e7-4d8f-96c9-421de725ab31
* Improve testability skill reliability
Refine routing and execution contracts from exact losing transcripts, strengthen behavioral eval checks, and add isolated C# fixtures without increasing repeated runs.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 84f7c88c-c8e7-4d8f-96c9-421de725ab31
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 84f7c88c-c8e7-4d8f-96c9-421de725ab31
2026-08-27 08:23:58 +00:00
f71cddd4a7
Allow Skill Value load retries
...
Keep one in-flight initialization request, retry failed index loads, and clarify why the derived index is excluded from plugin tabs.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5b4ee7ec-7244-4668-a151-5abee0ef2c4a
2026-08-27 00:13:41 -07:00
755e700f2c
Show n/a for null delta percent in Skill Value cells
...
When the baseline metric is 0, reduction() is null and the Tokens/Time delta
cell rendered a blank percent label. Show an explicit n/a instead, matching the
rollup cells, so the delta is never an ambiguous blank.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: e23e5efe-713f-4898-a158-8970f70fdcaf
2026-08-27 00:06:16 -07:00
86c1bba0f8
Merge branch 'main' into abhitejjohn-skill-value-dashboard
2026-08-26 23:43:26 -07:00
a3cb4a2a9f
Merge pull request #1045 from dotnet/tannergooding-port-vectorization-skill
...
Add general .NET vectorization skill
2026-08-26 13:26:33 -07:00
8bc7737ed9
Publish compact Skill Value dashboard index
...
Generate a retained skill-value.json index from per-plugin histories so the default dashboard tab loads only the compact value data instead of every full benchmark file. Keep derived data out of plugin tabs and harden purge enumeration.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 5b4ee7ec-7244-4668-a151-5abee0ef2c4a
2026-08-26 09:16:29 -07:00
3cd1337923
Improve dotnet test skill routing and quality ( #1051 )
...
* Improve coverage and tagging routing
Clarify two-sided sibling boundaries for coverage arithmetic, branch gaps, trait classification, and report-only frameworks. Replace prompt-echo graders while preserving outcome rubrics.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Address routing review feedback
Make coverage and tagging boundaries reciprocal, preserve convention-based Go tagging, and make outcome graders robust to multiline answers and equivalent uncovered wording.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Harden routing and grader equivalence
Preserve key anti-pattern triggers, complete coverage-prioritization handoffs, generalize convention-based tagging, and accept natural multiline and semantic grader equivalents.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Bound coverage outcome graders
Prevent substring false positives, accept equivalent branch and zero-coverage wording, preserve multiline Go classification, and restore the migration routing boundary.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Clarify project coverage outcomes
Scope hotspot thresholds to project-wide coverage, preserve behavioral threshold routing, and accept equivalent arithmetic and classification wording without weakening deterministic checks.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Tie graders to evaluated outcomes
Scope prioritization by project versus named target, require arithmetic context for 34 lines, preserve member identity checks, and bind Go semantic alternatives to divide-by-zero.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Refine CRAP routing and eval graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Tighten coverage evaluation contracts
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Bind coverage graders to evidence
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Close remaining eval grader gaps
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Use positive coverage grader assertions
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* De-cue semantic outcome graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Address routing review feedback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Scope anti-pattern coverage handoff
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Complete test routing boundaries
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Harden coverage and tagging graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Bound coverage grader literals
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Complete coverage routing constraints
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Require coverage and taxonomy evidence
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Enforce report-only fixture scope
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Move coverage arithmetic to rubric
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Fix coverage routing and table graders
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Add reciprocal routing guards
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Test tagging handoffs in sibling skills
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Tighten coverage and distribution boundaries
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Harden routing and report shape evals
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Resolve final coverage routing overlaps
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Require Go test rows in reports
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Improve cross-model test skill quality
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
* Remove cue-sensitive anti-pattern grader
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bfd14b28-8b12-4a2a-8735-bf4e91865e14
2026-08-26 12:33:27 +00:00
de3123d8a5
Fix agentic workflow runtime resolution ( #1062 )
...
* Fix agentic workflow runtime resolution
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Fix groom safe-output handling (#1063 )
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-26 12:20:57 +00:00
14d805d08f
Improve template skill eval coverage ( #1068 )
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-26 14:20:54 +02:00
fb3d055a49
Improve .NET test project scaffolding quality ( #1058 )
...
* Improve .NET test project scaffolding quality
Clarify create, repair, and no-op routing; broaden the eval with loss-tolerant existing-project scenarios while reducing repeated-run cost.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Improve test skill evaluation reliability
Fix natural-language routing gaps and scaffold completion behavior. Replace process-coupled code-testing graders with outcome checks, add predeclared TypeScript breadth, and reduce repeated-run cost.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 06f20828-94e0-4f48-aaff-6a2f46103d23
* Verify solution-level test discovery
Run the existing-project registration grader through Catalog.sln so it proves the CI entry point, not only direct project execution.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 06f20828-94e0-4f48-aaff-6a2f46103d23
* Improve test skill evaluation discrimination
Strengthen focused and broad test generation decisions, avoid xUnit v3 template churn, and add objective next-experiment breadth.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 06f20828-94e0-4f48-aaff-6a2f46103d23
* Fix scaffold workspace discovery
Use the task workspace rather than the skill installation directory, preserve complete verification evidence, and retire the unproven adjacent test-generation experiment.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 06f20828-94e0-4f48-aaff-6a2f46103d23
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: 06f20828-94e0-4f48-aaff-6a2f46103d23
2026-08-26 10:52:35 +02:00
3adf7cbc91
Improve run-tests routing and eval coverage ( #1052 )
...
* Improve run-tests routing and eval coverage
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Address run-tests review feedback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Clarify no-build test routing
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Address remaining run-tests review feedback
Tighten deterministic eval guards, correct MTP fixture signals, and clarify runner-specific guidance so validator and routing behavior match the documented contract.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Tighten run-tests evaluation scenarios
Make diagnostic and dump graders command-specific, keep advisory scenarios tool-free, and align routing and reference guidance with the actual consumer graph.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Cover remaining run-tests routing paths
Add no-tool no-build, bridged NUnit, and coverage scenarios; enforce direct argument placement and make the imported-props MTP fixture executable.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Address post-merge eval review feedback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Align run-tests and hot-reload review coverage
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Harden test runner boundaries and fixtures
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Complete test routing eval fixtures
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Align advisory routing assumptions
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Cover incomplete test bridge configurations
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Guard read-only platform evaluation
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Reduce test skill eval overfitting
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Scope VSTest TRX negative grader
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Tighten test skill response contracts
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Reject positional native MTP targets
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Focus platform detection responses
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Progressively disclose test command mode
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Require TUnit executable fallback
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
* Harden test routing from rerun evidence
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: bde0fbcd-b442-4ab7-b628-4beb1574a597
2026-08-26 08:48:45 +00:00
542d376c35
Address second review pass on Skill Value view
...
- Drill-down paired-observation count now uses min(baseline.n, treatment.n),
matching the table's paired-n, so inconsistent or legacy entries cannot
overstate the paired sample size.
- signedPct helper renders an exact-or-rounded zero reduction as "0%" with no
sign, so "no change" no longer reads as a "+0%" regression (deltaCell and the
rollup percent both use it).
- When skill-value.js is present but there are no evaluation plugins, fall back
to the Token Usage tab instead of leaving the page on the Skill Value panel's
permanent "Loading...".
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: e23e5efe-713f-4898-a158-8970f70fdcaf
2026-08-26 00:02:04 -07:00
98a6816daa
Improve test gap analysis routing and efficiency ( #1048 )
...
* Improve test gap analysis efficiency
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Clarify polyglot test gap routing
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Fix mutation catalog markdown lint
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Render logical OR correctly in mutation table
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Correct test gap analysis selection
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Clarify test gap request classification
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Distinguish static mutation candidates
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Improve test gap analysis consistency
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Fix logical OR table rendering
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Clarify unverified mutation candidates
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Name reportable mutation gap states
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Improve mutation gap verification consistency
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Clarify test gap evidence states
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Align test gap routing markers
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Align test writing gap criteria
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
* Clarify discovery guidance wording
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: db4db104-41a3-47d9-afba-24e260939c48
2026-08-26 06:54:27 +00:00
030ca2bba1
Merge pull request #1066 from dotnet/dependabot/github_actions/github-actions-dependencies-7a5a078ad4
...
Bump actions/checkout from 7.0.0 to 7.0.1 in the github-actions-dependencies group
2026-08-25 23:34:09 -07:00
2ca86e9d60
Merge pull request #1065 from dotnet/dependabot/npm_and_yarn/eng/evaluation-tools/microsoft/vally-cli-0.14.0
...
Bump @microsoft/vally-cli from 0.13.0 to 0.14.0 in /eng/evaluation-tools
2026-08-25 23:33:34 -07:00
b9aca2d25b
Bump actions/checkout in the github-actions-dependencies group
...
Bumps the github-actions-dependencies group with 1 update: [actions/checkout](https://github.com/actions/checkout ).
Updates `actions/checkout` from 7.0.0 to 7.0.1
- [Release notes](https://github.com/actions/checkout/releases )
- [Changelog](https://github.com/actions/checkout/blob/main/CHANGELOG.md )
- [Commits](https://github.com/actions/checkout/compare/v7...3d3c42e5aac5ba805825da76410c181273ba90b1 )
---
updated-dependencies:
- dependency-name: actions/checkout
dependency-version: 7.0.1
dependency-type: direct:production
update-type: version-update:semver-patch
dependency-group: github-actions-dependencies
...
Signed-off-by: dependabot[bot] <support@github.com >
2026-08-26 05:44:57 +00:00
3774d08eb3
Bump @microsoft/vally-cli from 0.13.0 to 0.14.0 in /eng/evaluation-tools
...
Bumps @microsoft/vally-cli from 0.13.0 to 0.14.0.
---
updated-dependencies:
- dependency-name: "@microsoft/vally-cli"
dependency-version: 0.14.0
dependency-type: direct:production
update-type: version-update:semver-minor
...
Signed-off-by: dependabot[bot] <support@github.com >
2026-08-26 05:44:37 +00:00
1daa97619f
Address final review on Skill Value view
...
- Load skill-value.js (and token-usage.js) before dashboard.js so
window.initSkillValue exists when the default Skill Value tab renders on
first load; previously the fallback activated the first plugin tab instead.
- Aggregate key now includes the plugin, so two plugins that share a skill
name no longer blend histories (the view is grouped Plugin -> Skill -> Model).
- valueSentence distinguishes "not-passed rate still gathering data" (pass
checks present but below the gate) from "No pass/fail data" (no boolean
graders), and describes the unit as counted trials rather than scenarios.
- Document that only the primary judge (vally-results-*) feeds benchmark data;
the dual-judge re-score (vally-crossjudge-*) feeds only judge-comparison.json,
so Skill Value and Quality/Efficiency both reflect the primary judge.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: e23e5efe-713f-4898-a158-8970f70fdcaf
2026-08-25 17:28:47 -07:00
2715e04578
Add per-skill "Skill Value" view to the evaluation dashboard
...
Adds a Plugin -> Skill -> Model "Skill Value" tab that surfaces, per skill,
the token and time a skill saves and how often scenarios pass without it.
Reuses the existing publish-eval-data job by emitting an entries.SkillValue
block from generate-benchmark-data.ps1 (no new job, no adapter change).
Fixes in this change:
- Quality/Efficiency stay one entry per executor model per run. The scheduled
benchmark merge now groups by (model, judge) for Skill Value, but passes
-SkillValueOnly for the second+ judge of a model so dual-judge runs no longer
duplicate same-model Quality/Efficiency points.
- Pass telemetry is labelled honestly. adapt.mjs notes the per-arm pass booleans
may include LLM-grader results, so the field is hasPassData (not
hasDeterministic), the column reads "Not-passed", and the claim reads "do not
pass their checks" with a tooltip caveat rather than an objective failure gate.
- Single-model groups show that model's real deltas; multi-model rollups read
"N model/judge result(s) show measured value" instead of an ambiguous 0/m.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
Copilot-Session: e23e5efe-713f-4898-a158-8970f70fdcaf
2026-08-25 17:26:55 -07:00
d4bf2bbf62
Clarify framework reuse and managed references
...
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-25 14:53:55 -07:00
1b896e91fe
Show authoritative verdict evidence in evaluation dashboard ( #1053 )
...
* Improve dashboard verdict evidence
Distinguish authoritative stimulus-vote gate evidence from score triage, preserve activation intent and judge excerpts, and keep legacy dashboard history compatible.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
* Use portable dashboard skill paths
Build reference-skill paths from Join-Path segments so the generator does not embed platform-specific separators.
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
---------
Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com >
2026-08-25 18:14:31 +00:00
Abhitej John