[skills-eval] dotnet-blazor: 9 skills, 36% pass — 1 P0, 1 P1, 5 to strengthen, 2 keep

[AbhitejJohn]

🔄 EDITED 2026-07-19 — correction per #909

This finding has been revised. The errored trials that drove the FIX-RELIABILITY (P0) flag(s) below were shown in #909 to be judge-side infrastructure failures — a disabled/throttled PAT (CAPIError 400 organization disabled) and Vally's session.idle judge timeout — not fixture nondeterminism. Verdict scores already exclude errored trials, so re-running the classifier with those judge errors dropped reclassifies the affected skill(s):

Skill	Was	Now
configure-auth	FIX-RELIABILITY (P0)	EFFICIENT-WIN (keep)

Struck-through text below is the superseded finding — in particular the "pin fixture SDK/tool versions" remediation does not apply. The title's P0 count has been updated. No eval re-run is required to correct the scores; recovering the lost judgments (optional) means re-judging those trials, not re-running the skill.

Context — cross-family skill evaluation: dotnet-blazor

This issue is self-contained: it captures everything a skill author needs to act on the dotnet-blazor plugin without opening the full report.

What this measures. Every runnable skill in dotnet/skills was run through Vally (0.7) on a cross-family matrix: 5 executor model families — opus-4.8, gpt-5.5, sonnet-4.6, haiku-4.5, mai-flash — each judged by a different family (judge ≠ executor; default judge = latest Opus, or GPT when Opus is the executor). For every skill, a skilled run is compared against a baseline (no-skill) run and scored per executor. This removes single-model and self-judging bias, so a skill that only helps one model family — or only its own family's judge — is visible.

• Data source: cross-family CI grid (run 29228914412 + backfills) — 5 executors × 85 runnable skills, 419 scored cells over 84 skills.
• Row grain: one row per skill, aggregated across its (up to 5) executor cells. avgN is the mean trial count behind the cells (trials 1–17; higher = more statistically trustworthy). thin-N flags directional-only rows.

dotnet-blazor at a glance (portfolio scorecard)

Plugin	Skills	Cells	Pass	Impact	Tie-trials	Err	avg ΔTok	avg ΔTurns	avg ΔTools	Headline
dotnet-blazor	9	45	36%	0.368	18	1	+42,425	+0.78	+1.53	Costly (plan-ui-change +408k); mixed

Insights. 9 skill(s); mean impact 0.37; 3/9 help ≥1 frontier model. No frontier-validated exemplar. Weak-model-only wins (frontier misses): use-js-interop. Weakest: support-prerendering (0.12).

Address first: configure-auth — reliability; support-prerendering — regression; plan-ui-change — cost.

Address first: support-prerendering — regression; plan-ui-change — cost. (configure-auth reclassified out of P0 — see EDIT above.)

How to read the table

Each skill is scored skilled vs. baseline on these axes:

Signal	Column	What it means	Good
Breadth	Families ✓ (n/5)	How many of the 5 model families the skill helps (per-family pass count, max 5/5)	4–5 / 5
Where	Passed on	Which families passed. Frontier = latest Opus + latest GPT are bold; Sonnet 4.6 / Haiku 4.5 / MAI Flash are mid/low-weight	frontier ✓
Magnitude	Impact (−1…+1)	How strongly the judge prefers skilled over baseline	≥ 0.4
Decisiveness	Ties▵	Trials where the judge saw no difference → skill is inert	low
Safety	Loss▵	Trials where skilled was WORSE than baseline → skill misfires	~0
Reliability	Err	Trials that errored/crashed in setup or judging	0
Efficiency	ΔTok / ΔTurns / ΔTools	Extra tokens / agent turns / tool calls vs baseline	≤ 0
Confidence	avgN	Mean trials behind the verdict; low N = directional only	≥ 3
Invocation	Call%	Share of skilled trials where the model actually invoked the skill	~100%

Families ✓ is cell-level (max 5/5); Ties▵/Loss▵ are trial-level tallies summed across all families (including the ones where the skill failed). A high Families ✓ next to non-zero Loss▵ is not a contradiction — see Passed on and the Action text for where losses landed.

Action buckets (each skill has one primary action; [flags] note secondary concerns):

Bucket	Priority	Meaning
FIX-RELIABILITY	🔴 P0	Errored trials / no verdict — stabilize the harness before trusting the score
FIX-DISCOVERY	🔴 P0	Model doesn't invoke it (Call% < 50%) — a triggering/description problem
FIX-REGRESSION	🔴 P0	Skilled is worse than baseline on many trials — the skill misfires
ADD-DECISIVENESS	🟠 P1	Called ~100% but ties dominate, ~0 impact — inert; needs sharper behavioral steps
TRIM-COST	🟠 P1	Passes but with heavy token/turn overhead — trim verbosity
EXEMPLAR	🟢 keep	Broad, strong, reliable win — use as a template
EFFICIENT-WIN	🟢 protect	Wins and cuts turns/tools — the ideal shape
KEEP-POLISH	🟢	Solid majority win; minor polish + more trials
STRENGTHEN	🟡 P2	Marginal/mixed lift — sharpen triggers & success criteria

Per-skill actions

Skill	Families ✓	Passed on (frontier bold)	Impact	Ties▵	Loss▵	Err	avgN	Call%	ΔTok	ΔTurns	ΔTools	Action
plan-ui-change	4/5	GPT, Sonnet, Haiku, MAI	0.54	1	1	0	5	100%	+407,812	+8.68	+10.24	TRIM-COST · 🟠 Passes but heavy (+407,812 tok). Value is real; cost isn't justified. Trim: (1) cut redundant/whole-file reads — point to specific sections; (2) replace "explore everything" with a targeted checklist; (3) move deep reference material behind links instead of inlining it. ([frontier-miss], [costly])
coordinate-components	2/5	GPT, Sonnet	0.48	2	1	0	2	100%	+121,041	+3.3	+3.7	STRENGTHEN · 🟡 Marginal/mixed (passes 2/5 — GPT, Sonnet). Diagnosis: mostly ties — too generic/non-prescriptive. Try: (1) sharpen the trigger so it fires only where it wins; (2) add 1–2 opinionated, concrete steps that change behaviour; (3) add trials to separate signal from noise. ([frontier-miss], [frontier-regressed], [costly])
create-blazor-project	0/5	—	0.45	0	2	0	3	100%	-392,992	-6.93	-4.47	STRENGTHEN · 🟡 Marginal/mixed (passes 0/5). Diagnosis: misses both frontier models — likely assumes context they solve unaided. Try: (1) sharpen the trigger so it fires only where it wins; (2) add 1–2 opinionated, concrete steps that change behaviour; (3) add trials to separate signal from noise. If frontier models never benefit, scope it explicitly to weaker models or reconsider its value. ([both-frontier-miss])
configure-auth	2/5	Haiku, MAI	0.44	3	1	1	1.8	100%	+92,263	-0.9	-1.3	FIX-RELIABILITY · 🔴 1 errored trial(s) — verdict can't be trusted until setup is deterministic. This is an infra/harness fix, not a content one: (1) capture the failing trial's stderr; (2) pin tool/SDK versions in the fixture; (3) add a setup smoke-check before scoring. ([reliability], [both-frontier-miss], [frontier-regressed], [efficient]) 🔄 EDIT (see #909): EFFICIENT-WIN · ✅ Better and cheaper — wins on Haiku, MAI while saving 0.9 turns / 1.3 tools (weak-model win; both frontier models miss/regress, so don't template blindly). The 1 errored trial was a judge-side session.idle timeout (haiku), not fixture nondeterminism. ([both-frontier-miss], [frontier-regressed], [efficient])
author-component	2/5	GPT, MAI	0.43	2	0	0	3.8	100%	+78,872	+3.46	+4.39	STRENGTHEN · 🟡 Marginal/mixed (passes 2/5 — GPT, MAI). Diagnosis: mostly ties — too generic/non-prescriptive. Try: (1) sharpen the trigger so it fires only where it wins; (2) add 1–2 opinionated, concrete steps that change behaviour; (3) add trials to separate signal from noise. ([frontier-miss])
use-js-interop	2/5	Sonnet, Haiku	0.38	2	2	0	3.8	85%	-20,440	-1.17	-1.57	EFFICIENT-WIN · ✅ Better and cheaper. Wins on Sonnet, Haiku while saving 1.17 turns, 1.57 tools, 20,440 tok. What's good: it adds value without bloat — the ideal shape. Protect the brevity; don't let it grow. ⚠️ But both frontier models miss — this may only be helping mid/low-weight models. Treat as a weak-model win, not a universal one; check the frontier trajectories before templating. ([both-frontier-miss], [efficient])
collect-user-input	2/5	Sonnet, MAI	0.24	2	1	0	2	90%	+55,008	+0.5	+1.4	STRENGTHEN · 🟡 Marginal/mixed (passes 2/5 — Sonnet, MAI). Diagnosis: mostly ties — too generic/non-prescriptive. Try: (1) sharpen the trigger so it fires only where it wins; (2) add 1–2 opinionated, concrete steps that change behaviour; (3) add trials to separate signal from noise. If frontier models never benefit, scope it explicitly to weaker models or reconsider its value. ([both-frontier-miss], [frontier-regressed])
fetch-and-send-data	1/5	Sonnet	0.22	4	1	0	2	70%	+87,601	+3.7	+5.7	STRENGTHEN · 🟡 Marginal/mixed (passes 1/5 — Sonnet). Diagnosis: mostly ties — too generic/non-prescriptive. Try: (1) sharpen the trigger so it fires only where it wins; (2) add 1–2 opinionated, concrete steps that change behaviour; (3) add trials to separate signal from noise. If frontier models never benefit, scope it explicitly to weaker models or reconsider its value. ([both-frontier-miss], [frontier-regressed])
support-prerendering	1/5	Haiku	0.12	2	4	0	2	100%	-47,342	-3.6	-4.3	FIX-REGRESSION · 🔴 Regresses ~40% of trials (worse than baseline), losing on GPT, Sonnet. The skill is over-applying. Fixes: (1) add explicit stop-conditions ("do NOT act when…"); (2) narrow the trigger to the exact scenario it helps; (3) demote prescriptive edits to suggestions the agent can decline. Prioritise the frontier miss (Opus, GPT). ([both-frontier-miss], [frontier-regressed])

---

_{Generated from the cross-family Call-to-Action report (CALL-TO-ACTION.md §4–§5; companion IMPACT-ANALYSIS.md). Regenerate the underlying tables with node deep-metrics.mjs "$env:TEMP\cf-ci" agg-ci → node gen-cta-tables.mjs agg-ci. Numbers are directional where avgN is low; treat single-trial cells as hypotheses to confirm with more runs.}

[github-actions]

🎯 Issue Triage

Summary: This is an automated skill evaluation report for the dotnet-blazor plugin, generated from a cross-family CI grid run. It covers 9 skills with a 36% pass rate, identifying 2 P0 items (configure-auth reliability, support-prerendering regression), 1 P1 (plan-ui-change cost), 5 skills to strengthen, and 1 efficient win. This falls under the Blazor/ASP.NET Core area.

Type: task — automated action items for skill authors to address.

Assigned owners: @webreidi, @AbhitejJohn (defaults). The primary CODEOWNERS team @dotnet/aspnet should also be engaged for /plugins/dotnet-blazor/.

Confidence: High — clear area mapping, type is unambiguous, owners are well-defined.

Similar issues: No distinct duplicates found; this appears to be a unique evaluation run report.

Triage details

• Area mapping: area-blazor (already present) — maps to /plugins/dotnet-blazor/ per CODEOWNERS → @dotnet/aspnet
• CODEOWNERS path match: /plugins/dotnet-blazor/ @dotnet/aspnet``
• Confidence rationale: Issue is machine-generated with a clear structure; area and type are unambiguous. Teams cannot be set as assignees, so default repo owners were assigned instead.
• Suggested next steps: The @dotnet/aspnet team should prioritize configure-auth (FIX-RELIABILITY 🔴) and support-prerendering (FIX-REGRESSION 🔴) before addressing P1/P2 items.

Generated by Issue Triage for issue #891 · 68.4 AIC · ⊞ 28.4K · ◷

[AbhitejJohn]

Verifying fixes for this plugin's skills

To re-run the cross-family evaluation and confirm a change moves the pass rate, cherry-pick the dispatch workflow from #920 and dispatch it scoped to this plugin:

git fetch origin abhitejjohn-cross-family-eval-workflow
git cherry-pick 7c97b7346          # single commit in #920
git push origin HEAD

gh workflow run evaluation-run.yml --ref <your-branch> \
  -f plugins="dotnet-blazor" \
  -f executors="opus,gpt,sonnet46,haiku,mai" -f runs=3

To verify a single skill faster, scope with skills (the skill directory name, e.g. one of the skills listed above):

gh workflow run evaluation-run.yml --ref <your-branch> \
  -f skills="<skill-dir>" -f executors="opus,gpt" -f runs=3

Each skill is run skilled-vs-baseline across the 5-family executor matrix, with the judge never in the executor's family (default judge = latest Opus; GPT when Opus executes). Results assemble into a cross-family pass matrix you can compare against the numbers in this issue. Full inputs/usage: #920.