Adding keyboard support

[mgifford]

Added comprehensive keyboard accessibility instructions for GitHub Copilot, detailing WCAG 2.2 AA patterns for focus management, dialog traps, and more.

Pull Request Checklist

• I have read and followed the CONTRIBUTING.md guidelines.
• I have read and followed the Guidance for submissions involving paid services.
• My contribution adds a new instruction, prompt, agent, skill, or workflow file in the correct directory.
• The file follows the required naming convention.
• The content is clearly structured and follows the example format.
• I have tested my instructions, prompt, agent, skill, or workflow with GitHub Copilot.
• I have run npm start and verified that README.md is up to date.
• I am targeting the staged branch for this pull request.

---

Description

---

Type of Contribution

• New instruction file.
• New prompt file.
• New agent file.
• New plugin.
• New skill file.
• New agentic workflow.
• Update to existing instruction, prompt, agent, plugin, skill, or workflow.
• Other (please specify):

---

Additional Notes

---

By submitting this pull request, I confirm that my contribution abides by the Code of Conduct and will be licensed under the MIT License.

[github-actions]

🔍 Skill Validator Results

⛔ Findings need attention

Scope	Checked
Skills	132
Agents	1
Total	133
Severity	Count
---	---:
❌ Errors	19
⚠️ Warnings	126
ℹ️ Advisories	0

Summary

Level	Finding
❌	[geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	[geofeed-tuner] File reference 'assets/example/01-user-input-rfc8805-feed.csv' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	[geofeed-tuner] File reference './run/data/comments.json' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	[geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	[geofeed-tuner] File reference './run/data/mcp-server-payload.json' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	[geofeed-tuner] File reference './run/data/mcp-server-payload.json' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	[geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	[geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md.
❌	Duplicate skill name 'azure-resource-health-diagnose' found in 'plugins/devops-oncall/skills/azure-resource-health-diagnose' (first seen in 'plugins/azure-cloud-development/skills/azure-resource-health-diagnose')
❌	Duplicate skill name 'openapi-to-application-code' found in 'plugins/openapi-to-application-go/skills/openapi-to-application-code' (first seen in 'plugins/openapi-to-application-csharp-dotnet/skills/openapi-to-application-code')

Full validator output

```text Found 132 skill(s) [arize-ai-provider-integration] 📊 arize-ai-provider-integration: 2,305 BPE tokens [chars/4: 2,216] (detailed ✓), 29 sections, 16 code blocks [arize-ai-provider-integration] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [arize-annotation] 📊 arize-annotation: 1,822 BPE tokens [chars/4: 1,930] (detailed ✓), 18 sections, 7 code blocks [arize-annotation] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [arize-dataset] 📊 arize-dataset: 3,482 BPE tokens [chars/4: 3,444] (standard ~), 51 sections, 16 code blocks [arize-dataset] ⚠ Skill is 3,482 BPE tokens (chars/4 estimate: 3,444) — approaching "comprehensive" range where gains diminish. [arize-evaluator] 📊 arize-evaluator: 6,243 BPE tokens [chars/4: 6,326] (comprehensive ✗), 57 sections, 22 code blocks [arize-evaluator] ⚠ Skill is 6,243 BPE tokens (chars/4 estimate: 6,326) — "comprehensive" skills hurt performance by 2.9pp on average. Consider splitting into 2–3 focused skills. [arize-experiment] 📊 arize-experiment: 3,204 BPE tokens [chars/4: 3,167] (standard ~), 33 sections, 18 code blocks [arize-experiment] ⚠ Skill is 3,204 BPE tokens (chars/4 estimate: 3,167) — approaching "comprehensive" range where gains diminish. [arize-instrumentation] 📊 arize-instrumentation: 4,790 BPE tokens [chars/4: 4,901] (standard ~), 19 sections, 3 code blocks [arize-instrumentation] ⚠ Skill is 4,790 BPE tokens (chars/4 estimate: 4,901) — approaching "comprehensive" range where gains diminish. [arize-link] 📊 arize-link: 1,191 BPE tokens [chars/4: 1,073] (detailed ✓), 9 sections, 6 code blocks [arize-prompt-optimization] 📊 arize-prompt-optimization: 4,296 BPE tokens [chars/4: 4,594] (standard ~), 58 sections, 19 code blocks [arize-prompt-optimization] ⚠ Skill is 4,296 BPE tokens (chars/4 estimate: 4,594) — approaching "comprehensive" range where gains diminish. [arize-trace] 📊 arize-trace: 5,220 BPE tokens [chars/4: 5,151] (comprehensive ✗), 42 sections, 10 code blocks [arize-trace] ⚠ Skill is 5,220 BPE tokens (chars/4 estimate: 5,151) — "comprehensive" skills hurt performance by 2.9pp on average. Consider splitting into 2–3 focused skills. [automate-this] 📊 automate-this: 2,869 BPE tokens [chars/4: 3,107] (standard ~), 16 sections, 8 code blocks [automate-this] ⚠ Skill is 2,869 BPE tokens (chars/4 estimate: 3,107) — approaching "comprehensive" range where gains diminish. [suggest-awesome-github-copilot-agents] 📊 suggest-awesome-github-copilot-agents: 1,467 BPE tokens [chars/4: 1,979] (detailed ✓), 9 sections, 0 code blocks [suggest-awesome-github-copilot-agents] ⚠ No code blocks — agents perform better with concrete snippets and commands. [suggest-awesome-github-copilot-instructions] 📊 suggest-awesome-github-copilot-instructions: 1,499 BPE tokens [chars/4: 1,904] (detailed ✓), 11 sections, 1 code blocks [suggest-awesome-github-copilot-skills] 📊 suggest-awesome-github-copilot-skills: 1,725 BPE tokens [chars/4: 2,018] (detailed ✓), 11 sections, 1 code blocks [az-cost-optimize] 📊 az-cost-optimize: 2,956 BPE tokens [chars/4: 3,218] (standard ~), 12 sections, 7 code blocks [az-cost-optimize] ⚠ Skill is 2,956 BPE tokens (chars/4 estimate: 3,218) — approaching "comprehensive" range where gains diminish. [azure-pricing] 📊 azure-pricing: 2,315 BPE tokens [chars/4: 2,371] (detailed ✓), 23 sections, 4 code blocks [azure-resource-health-diagnose] 📊 azure-resource-health-diagnose: 2,492 BPE tokens [chars/4: 2,859] (detailed ✓), 12 sections, 8 code blocks [import-infrastructure-as-code] 📊 import-infrastructure-as-code: 4,135 BPE tokens [chars/4: 4,680] (standard ~), 30 sections, 11 code blocks [import-infrastructure-as-code] ⚠ Skill is 4,135 BPE tokens (chars/4 estimate: 4,680) — approaching "comprehensive" range where gains diminish. [remember-interactive-programming] 📊 remember-interactive-programming: 175 BPE tokens [chars/4: 221] (compact ✓), 0 sections, 0 code blocks [remember-interactive-programming] ⚠ Skill is only 175 BPE tokens (chars/4 estimate: 221) — may be too sparse to provide actionable guidance. [remember-interactive-programming] ⚠ No section headers — agents navigate structured documents better. [remember-interactive-programming] ⚠ No code blocks — agents perform better with concrete snippets and commands. [remember-interactive-programming] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [context-map] 📊 context-map: 249 BPE tokens [chars/4: 287] (compact ✓), 10 sections, 1 code blocks [refactor-plan] 📊 refactor-plan: 356 BPE tokens [chars/4: 355] (compact ✓), 15 sections, 1 code blocks [what-context-needed] 📊 what-context-needed: 209 BPE tokens [chars/4: 219] (compact ✓), 9 sections, 1 code blocks [integrate-context-matic] 📊 integrate-context-matic: 1,412 BPE tokens [chars/4: 1,495] (detailed ✓), 12 sections, 0 code blocks [integrate-context-matic] ⚠ No code blocks — agents perform better with concrete snippets and commands. [integrate-context-matic] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [onboard-context-matic] 📊 onboard-context-matic: 3,039 BPE tokens [chars/4: 3,261] (standard ~), 13 sections, 12 code blocks [onboard-context-matic] ⚠ Skill is 3,039 BPE tokens (chars/4 estimate: 3,261) — approaching "comprehensive" range where gains diminish. [copilot-sdk] 📊 copilot-sdk: 5,749 BPE tokens [chars/4: 6,042] (comprehensive ✗), 63 sections, 41 code blocks [copilot-sdk] ⚠ Skill is 5,749 BPE tokens (chars/4 estimate: 6,042) — "comprehensive" skills hurt performance by 2.9pp on average. Consider splitting into 2–3 focused skills. [aspnet-minimal-api-openapi] 📊 aspnet-minimal-api-openapi: 343 BPE tokens [chars/4: 439] (compact ✓), 5 sections, 0 code blocks [aspnet-minimal-api-openapi] ⚠ No code blocks — agents perform better with concrete snippets and commands. [aspnet-minimal-api-openapi] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [csharp-async] 📊 csharp-async: 379 BPE tokens [chars/4: 466] (compact ✓), 7 sections, 0 code blocks [csharp-async] ⚠ No code blocks — agents perform better with concrete snippets and commands. [csharp-async] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [csharp-mstest] 📊 csharp-mstest: 3,249 BPE tokens [chars/4: 3,458] (standard ~), 31 sections, 22 code blocks [csharp-mstest] ⚠ Skill is 3,249 BPE tokens (chars/4 estimate: 3,458) — approaching "comprehensive" range where gains diminish. [csharp-nunit] 📊 csharp-nunit: 636 BPE tokens [chars/4: 711] (detailed ✓), 8 sections, 0 code blocks [csharp-nunit] ⚠ No code blocks — agents perform better with concrete snippets and commands. [csharp-nunit] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [csharp-tunit] 📊 csharp-tunit: 1,164 BPE tokens [chars/4: 1,287] (detailed ✓), 10 sections, 0 code blocks [csharp-tunit] ⚠ No code blocks — agents perform better with concrete snippets and commands. [csharp-tunit] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [csharp-xunit] 📊 csharp-xunit: 599 BPE tokens [chars/4: 677] (detailed ✓), 8 sections, 0 code blocks [csharp-xunit] ⚠ No code blocks — agents perform better with concrete snippets and commands. [csharp-xunit] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [dotnet-best-practices] 📊 dotnet-best-practices: 602 BPE tokens [chars/4: 808] (detailed ✓), 12 sections, 0 code blocks [dotnet-best-practices] ⚠ No code blocks — agents perform better with concrete snippets and commands. [dotnet-best-practices] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [dotnet-upgrade] 📊 dotnet-upgrade: 1,354 BPE tokens [chars/4: 1,768] (detailed ✓), 1 sections, 0 code blocks [dotnet-upgrade] ⚠ No code blocks — agents perform better with concrete snippets and commands. [dotnet-upgrade] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [csharp-mcp-server-generator] 📊 csharp-mcp-server-generator: 459 BPE tokens [chars/4: 560] (detailed ✓), 9 sections, 0 code blocks [csharp-mcp-server-generator] ⚠ No code blocks — agents perform better with concrete snippets and commands. [postgresql-code-review] 📊 postgresql-code-review: 1,646 BPE tokens [chars/4: 1,847] (detailed ✓), 21 sections, 8 code blocks [postgresql-optimization] 📊 postgresql-optimization: 2,651 BPE tokens [chars/4: 2,834] (standard ~), 36 sections, 20 code blocks [postgresql-optimization] ⚠ Skill is 2,651 BPE tokens (chars/4 estimate: 2,834) — approaching "comprehensive" range where gains diminish. [postgresql-optimization] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [sql-code-review] 📊 sql-code-review: 2,112 BPE tokens [chars/4: 2,220] (detailed ✓), 36 sections, 15 code blocks [sql-optimization] 📊 sql-optimization: 2,212 BPE tokens [chars/4: 2,298] (detailed ✓), 28 sections, 14 code blocks [dataverse-python-advanced-patterns] 📊 dataverse-python-advanced-patterns: 242 BPE tokens [chars/4: 291] (compact ✓), 0 sections, 0 code blocks [dataverse-python-advanced-patterns] ⚠ No section headers — agents navigate structured documents better. [dataverse-python-advanced-patterns] ⚠ No code blocks — agents perform better with concrete snippets and commands. [dataverse-python-production-code] 📊 dataverse-python-production-code: 813 BPE tokens [chars/4: 913] (detailed ✓), 9 sections, 3 code blocks [dataverse-python-quickstart] 📊 dataverse-python-quickstart: 133 BPE tokens [chars/4: 163] (compact ✓), 0 sections, 0 code blocks [dataverse-python-quickstart] ⚠ Skill is only 133 BPE tokens (chars/4 estimate: 163) — may be too sparse to provide actionable guidance. [dataverse-python-quickstart] ⚠ No section headers — agents navigate structured documents better. [dataverse-python-quickstart] ⚠ No code blocks — agents perform better with concrete snippets and commands. [dataverse-python-quickstart] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [dataverse-python-usecase-builder] 📊 dataverse-python-usecase-builder: 1,538 BPE tokens [chars/4: 1,684] (detailed ✓), 38 sections, 5 code blocks [azure-resource-health-diagnose] 📊 azure-resource-health-diagnose: 2,492 BPE tokens [chars/4: 2,859] (detailed ✓), 12 sections, 8 code blocks [multi-stage-dockerfile] 📊 multi-stage-dockerfile: 443 BPE tokens [chars/4: 577] (detailed ✓), 5 sections, 0 code blocks [multi-stage-dockerfile] ⚠ No code blocks — agents perform better with concrete snippets and commands. [multi-stage-dockerfile] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [doublecheck] 📊 doublecheck: 3,254 BPE tokens [chars/4: 3,973] (standard ~), 26 sections, 3 code blocks [doublecheck] ⚠ Skill is 3,254 BPE tokens (chars/4 estimate: 3,973) — approaching "comprehensive" range where gains diminish. [from-the-other-side-vega] 📊 from-the-other-side-vega: 820 BPE tokens [chars/4: 894] (detailed ✓), 9 sections, 0 code blocks [from-the-other-side-vega] ⚠ No code blocks — agents perform better with concrete snippets and commands. [from-the-other-side-vega] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [geofeed-tuner] 📊 geofeed-tuner: 11,662 BPE tokens [chars/4: 12,215] (comprehensive ✗), 35 sections, 15 code blocks ❌ [geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md. ❌ [geofeed-tuner] File reference 'assets/example/01-user-input-rfc8805-feed.csv' is 2 directories deep — maximum is 1 level from SKILL.md. ❌ [geofeed-tuner] File reference './run/data/comments.json' is 2 directories deep — maximum is 1 level from SKILL.md. ❌ [geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md. ❌ [geofeed-tuner] File reference './run/data/mcp-server-payload.json' is 2 directories deep — maximum is 1 level from SKILL.md. ❌ [geofeed-tuner] File reference './run/data/mcp-server-payload.json' is 2 directories deep — maximum is 1 level from SKILL.md. ❌ [geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md. ❌ [geofeed-tuner] File reference './run/data/report-data.json' is 2 directories deep — maximum is 1 level from SKILL.md. [geofeed-tuner] ⚠ Skill is 11,662 BPE tokens (chars/4 estimate: 12,215) — "comprehensive" skills hurt performance by 2.9pp on average. Consider splitting into 2–3 focused skills. [flowstudio-power-automate-build] 📊 flowstudio-power-automate-build: 4,461 BPE tokens [chars/4: 4,595] (standard ~), 40 sections, 15 code blocks [flowstudio-power-automate-build] ⚠ Skill is 4,461 BPE tokens (chars/4 estimate: 4,595) — approaching "comprehensive" range where gains diminish. [flowstudio-power-automate-build] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [flowstudio-power-automate-debug] 📊 flowstudio-power-automate-debug: 4,119 BPE tokens [chars/4: 4,060] (standard ~), 64 sections, 15 code blocks [flowstudio-power-automate-debug] ⚠ Skill is 4,119 BPE tokens (chars/4 estimate: 4,060) — approaching "comprehensive" range where gains diminish. [flowstudio-power-automate-governance] 📊 flowstudio-power-automate-governance: 5,229 BPE tokens [chars/4: 5,609] (comprehensive ✗), 17 sections, 11 code blocks [flowstudio-power-automate-governance] ⚠ Skill is 5,229 BPE tokens (chars/4 estimate: 5,609) — "comprehensive" skills hurt performance by 2.9pp on average. Consider splitting into 2–3 focused skills. [flowstudio-power-automate-mcp] 📊 flowstudio-power-automate-mcp: 4,716 BPE tokens [chars/4: 4,522] (standard ~), 62 sections, 11 code blocks [flowstudio-power-automate-mcp] ⚠ Skill is 4,716 BPE tokens (chars/4 estimate: 4,522) — approaching "comprehensive" range where gains diminish. [flowstudio-power-automate-mcp] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [flowstudio-power-automate-monitoring] 📊 flowstudio-power-automate-monitoring: 3,546 BPE tokens [chars/4: 3,354] (standard ~), 25 sections, 14 code blocks [flowstudio-power-automate-monitoring] ⚠ Skill is 3,546 BPE tokens (chars/4 estimate: 3,354) — approaching "comprehensive" range where gains diminish. [playwright-explore-website] 📊 playwright-explore-website: 140 BPE tokens [chars/4: 171] (compact ✓), 2 sections, 0 code blocks [playwright-explore-website] ⚠ Skill is only 140 BPE tokens (chars/4 estimate: 171) — may be too sparse to provide actionable guidance. [playwright-explore-website] ⚠ No code blocks — agents perform better with concrete snippets and commands. [playwright-generate-test] 📊 playwright-generate-test: 178 BPE tokens [chars/4: 220] (compact ✓), 2 sections, 0 code blocks [playwright-generate-test] ⚠ Skill is only 178 BPE tokens (chars/4 estimate: 220) — may be too sparse to provide actionable guidance. [playwright-generate-test] ⚠ No code blocks — agents perform better with concrete snippets and commands. [playwright-generate-test] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [go-mcp-server-generator] 📊 go-mcp-server-generator: 1,740 BPE tokens [chars/4: 1,759] (detailed ✓), 21 sections, 8 code blocks [create-spring-boot-java-project] 📊 create-spring-boot-java-project: 1,110 BPE tokens [chars/4: 1,159] (detailed ✓), 17 sections, 11 code blocks [create-spring-boot-java-project] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [java-docs] 📊 java-docs: 316 BPE tokens [chars/4: 360] (compact ✓), 1 sections, 0 code blocks [java-docs] ⚠ No code blocks — agents perform better with concrete snippets and commands. [java-docs] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [java-junit] 📊 java-junit: 695 BPE tokens [chars/4: 746] (detailed ✓), 8 sections, 0 code blocks [java-junit] ⚠ No code blocks — agents perform better with concrete snippets and commands. [java-junit] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [java-springboot] 📊 java-springboot: 855 BPE tokens [chars/4: 1,034] (detailed ✓), 10 sections, 0 code blocks [java-springboot] ⚠ No code blocks — agents perform better with concrete snippets and commands. [java-springboot] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [java-mcp-server-generator] 📊 java-mcp-server-generator: 4,637 BPE tokens [chars/4: 5,506] (standard ~), 28 sections, 19 code blocks [java-mcp-server-generator] ⚠ Skill is 4,637 BPE tokens (chars/4 estimate: 5,506) — approaching "comprehensive" range where gains diminish. [kotlin-mcp-server-generator] 📊 kotlin-mcp-server-generator: 2,456 BPE tokens [chars/4: 2,672] (detailed ✓), 29 sections, 13 code blocks [mcp-create-adaptive-cards] 📊 mcp-create-adaptive-cards: 2,949 BPE tokens [chars/4: 3,136] (standard ~), 33 sections, 22 code blocks [mcp-create-adaptive-cards] ⚠ Skill is 2,949 BPE tokens (chars/4 estimate: 3,136) — approaching "comprehensive" range where gains diminish. [mcp-create-declarative-agent] 📊 mcp-create-declarative-agent: 1,978 BPE tokens [chars/4: 2,036] (detailed ✓), 30 sections, 12 code blocks [mcp-deploy-manage-agents] 📊 mcp-deploy-manage-agents: 2,036 BPE tokens [chars/4: 2,471] (detailed ✓), 49 sections, 2 code blocks [napkin] 📊 napkin: 1,649 BPE tokens [chars/4: 1,784] (detailed ✓), 12 sections, 4 code blocks [noob-mode] 📊 noob-mode: 2,604 BPE tokens [chars/4: 2,765] (standard ~), 13 sections, 9 code blocks [noob-mode] ⚠ Skill is 2,604 BPE tokens (chars/4 estimate: 2,765) — approaching "comprehensive" range where gains diminish. [openapi-to-application-code] 📊 openapi-to-application-code: 931 BPE tokens [chars/4: 1,144] (detailed ✓), 11 sections, 1 code blocks [openapi-to-application-code] 📊 openapi-to-application-code: 931 BPE tokens [chars/4: 1,144] (detailed ✓), 11 sections, 1 code blocks [openapi-to-application-code] 📊 openapi-to-application-code: 931 BPE tokens [chars/4: 1,144] (detailed ✓), 11 sections, 1 code blocks [openapi-to-application-code] 📊 openapi-to-application-code: 931 BPE tokens [chars/4: 1,144] (detailed ✓), 11 sections, 1 code blocks [openapi-to-application-code] 📊 openapi-to-application-code: 931 BPE tokens [chars/4: 1,144] (detailed ✓), 11 sections, 1 code blocks [creating-oracle-to-postgres-master-migration-plan] 📊 creating-oracle-to-postgres-master-migration-plan: 755 BPE tokens [chars/4: 790] (detailed ✓), 6 sections, 2 code blocks [creating-oracle-to-postgres-migration-bug-report] 📊 creating-oracle-to-postgres-migration-bug-report: 473 BPE tokens [chars/4: 567] (detailed ✓), 6 sections, 0 code blocks [creating-oracle-to-postgres-migration-bug-report] ⚠ No code blocks — agents perform better with concrete snippets and commands. [creating-oracle-to-postgres-migration-bug-report] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [creating-oracle-to-postgres-migration-integration-tests] 📊 creating-oracle-to-postgres-migration-integration-tests: 605 BPE tokens [chars/4: 760] (detailed ✓), 4 sections, 1 code blocks [creating-oracle-to-postgres-migration-integration-tests] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [migrating-oracle-to-postgres-stored-procedures] 📊 migrating-oracle-to-postgres-stored-procedures: 493 BPE tokens [chars/4: 576] (detailed ✓), 2 sections, 1 code blocks [migrating-oracle-to-postgres-stored-procedures] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [planning-oracle-to-postgres-migration-integration-testing] 📊 planning-oracle-to-postgres-migration-integration-testing: 438 BPE tokens [chars/4: 547] (detailed ✓), 4 sections, 1 code blocks [planning-oracle-to-postgres-migration-integration-testing] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [reviewing-oracle-to-postgres-migration] 📊 reviewing-oracle-to-postgres-migration: 605 BPE tokens [chars/4: 762] (detailed ✓), 5 sections, 2 code blocks [scaffolding-oracle-to-postgres-migration-test-project] 📊 scaffolding-oracle-to-postgres-migration-test-project: 527 BPE tokens [chars/4: 632] (detailed ✓), 3 sections, 1 code blocks [scaffolding-oracle-to-postgres-migration-test-project] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [sponsor-finder] 📊 sponsor-finder: 3,085 BPE tokens [chars/4: 2,991] (standard ~), 25 sections, 6 code blocks [sponsor-finder] ⚠ Skill is 3,085 BPE tokens (chars/4 estimate: 2,991) — approaching "comprehensive" range where gains diminish. [phoenix-cli] 📊 phoenix-cli: 1,615 BPE tokens [chars/4: 1,563] (detailed ✓), 13 sections, 12 code blocks [phoenix-cli] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [phoenix-evals] 📊 phoenix-evals: 1,089 BPE tokens [chars/4: 1,126] (detailed ✓), 5 sections, 0 code blocks [phoenix-evals] ⚠ No code blocks — agents perform better with concrete snippets and commands. [phoenix-evals] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [phoenix-tracing] 📊 phoenix-tracing: 1,488 BPE tokens [chars/4: 1,690] (detailed ✓), 17 sections, 1 code blocks [php-mcp-server-generator] 📊 php-mcp-server-generator: 2,723 BPE tokens [chars/4: 2,697] (standard ~), 37 sections, 25 code blocks [php-mcp-server-generator] ⚠ Skill is 2,723 BPE tokens (chars/4 estimate: 2,697) — approaching "comprehensive" range where gains diminish. [polyglot-test-agent] 📊 polyglot-test-agent: 1,611 BPE tokens [chars/4: 1,676] (detailed ✓), 22 sections, 5 code blocks [power-apps-code-app-scaffold] 📊 power-apps-code-app-scaffold: 1,424 BPE tokens [chars/4: 1,786] (detailed ✓), 18 sections, 2 code blocks [power-bi-dax-optimization] 📊 power-bi-dax-optimization: 1,114 BPE tokens [chars/4: 1,390] (detailed ✓), 18 sections, 6 code blocks [power-bi-model-design-review] 📊 power-bi-model-design-review: 2,126 BPE tokens [chars/4: 2,782] (detailed ✓), 28 sections, 17 code blocks [power-bi-performance-troubleshooting] 📊 power-bi-performance-troubleshooting: 2,078 BPE tokens [chars/4: 2,591] (detailed ✓), 31 sections, 19 code blocks [power-bi-report-design-consultation] 📊 power-bi-report-design-consultation: 2,357 BPE tokens [chars/4: 2,723] (detailed ✓), 22 sections, 13 code blocks [mcp-copilot-studio-server-generator] 📊 mcp-copilot-studio-server-generator: 942 BPE tokens [chars/4: 1,063] (detailed ✓), 6 sections, 2 code blocks [power-platform-mcp-connector-suite] 📊 power-platform-mcp-connector-suite: 1,348 BPE tokens [chars/4: 1,590] (detailed ✓), 19 sections, 1 code blocks [power-platform-mcp-connector-suite] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [breakdown-epic-arch] 📊 breakdown-epic-arch: 601 BPE tokens [chars/4: 712] (detailed ✓), 12 sections, 0 code blocks [breakdown-epic-arch] ⚠ No code blocks — agents perform better with concrete snippets and commands. [breakdown-epic-arch] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [breakdown-epic-pm] 📊 breakdown-epic-pm: 509 BPE tokens [chars/4: 575] (detailed ✓), 13 sections, 0 code blocks [breakdown-epic-pm] ⚠ No code blocks — agents perform better with concrete snippets and commands. [breakdown-epic-pm] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [breakdown-feature-implementation] 📊 breakdown-feature-implementation: 966 BPE tokens [chars/4: 1,189] (detailed ✓), 9 sections, 2 code blocks [breakdown-feature-implementation] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [breakdown-feature-prd] 📊 breakdown-feature-prd: 557 BPE tokens [chars/4: 620] (detailed ✓), 13 sections, 0 code blocks [breakdown-feature-prd] ⚠ No code blocks — agents perform better with concrete snippets and commands. [breakdown-feature-prd] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [create-github-issues-feature-from-implementation-plan] 📊 create-github-issues-feature-from-implementation-plan: 191 BPE tokens [chars/4: 237] (compact ✓), 4 sections, 0 code blocks [create-github-issues-feature-from-implementation-plan] ⚠ Skill is only 191 BPE tokens (chars/4 estimate: 237) — may be too sparse to provide actionable guidance. [create-github-issues-feature-from-implementation-plan] ⚠ No code blocks — agents perform better with concrete snippets and commands. [create-implementation-plan] 📊 create-implementation-plan: 1,390 BPE tokens [chars/4: 1,575] (detailed ✓), 22 sections, 1 code blocks [create-implementation-plan] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [create-technical-spike] 📊 create-technical-spike: 1,358 BPE tokens [chars/4: 1,643] (detailed ✓), 33 sections, 1 code blocks [update-implementation-plan] 📊 update-implementation-plan: 1,400 BPE tokens [chars/4: 1,595] (detailed ✓), 22 sections, 1 code blocks [update-implementation-plan] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [python-mcp-server-generator] 📊 python-mcp-server-generator: 871 BPE tokens [chars/4: 990] (detailed ✓), 13 sections, 0 code blocks [python-mcp-server-generator] ⚠ No code blocks — agents perform better with concrete snippets and commands. [react-audit-grep-patterns] 📊 react-audit-grep-patterns: 475 BPE tokens [chars/4: 461] (detailed ✓), 13 sections, 1 code blocks [react-audit-grep-patterns] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [react18-batching-patterns] 📊 react18-batching-patterns: 561 BPE tokens [chars/4: 621] (detailed ✓), 4 sections, 1 code blocks [react18-batching-patterns] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [react18-dep-compatibility] 📊 react18-dep-compatibility: 1,388 BPE tokens [chars/4: 1,124] (detailed ✓), 10 sections, 1 code blocks [react18-dep-compatibility] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [react18-enzyme-to-rtl] 📊 react18-enzyme-to-rtl: 870 BPE tokens [chars/4: 917] (detailed ✓), 6 sections, 3 code blocks [react18-legacy-context] 📊 react18-legacy-context: 561 BPE tokens [chars/4: 620] (detailed ✓), 8 sections, 2 code blocks [react18-lifecycle-patterns] 📊 react18-lifecycle-patterns: 768 BPE tokens [chars/4: 910] (detailed ✓), 7 sections, 1 code blocks [react18-lifecycle-patterns] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [react18-string-refs] 📊 react18-string-refs: 549 BPE tokens [chars/4: 527] (detailed ✓), 6 sections, 1 code blocks [react19-concurrent-patterns] 📊 react19-concurrent-patterns: 681 BPE tokens [chars/4: 722] (detailed ✓), 9 sections, 5 code blocks [react19-source-patterns] 📊 react19-source-patterns: 462 BPE tokens [chars/4: 489] (detailed ✓), 4 sections, 1 code blocks [react19-source-patterns] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [react19-test-patterns] 📊 react19-test-patterns: 815 BPE tokens [chars/4: 790] (detailed ✓), 7 sections, 5 code blocks [roundup] 📊 roundup: 2,219 BPE tokens [chars/4: 2,591] (detailed ✓), 28 sections, 0 code blocks [roundup] ⚠ No code blocks — agents perform better with concrete snippets and commands. [roundup] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [roundup-setup] 📊 roundup-setup: 3,202 BPE tokens [chars/4: 3,669] (standard ~), 18 sections, 1 code blocks [roundup-setup] ⚠ Skill is 3,202 BPE tokens (chars/4 estimate: 3,669) — approaching "comprehensive" range where gains diminish. [ruby-mcp-server-generator] 📊 ruby-mcp-server-generator: 3,459 BPE tokens [chars/4: 3,631] (standard ~), 37 sections, 24 code blocks [ruby-mcp-server-generator] ⚠ Skill is 3,459 BPE tokens (chars/4 estimate: 3,631) — approaching "comprehensive" range where gains diminish. [rust-mcp-server-generator] 📊 rust-mcp-server-generator: 3,088 BPE tokens [chars/4: 3,190] (standard ~), 31 sections, 24 code blocks [rust-mcp-server-generator] ⚠ Skill is 3,088 BPE tokens (chars/4 estimate: 3,190) — approaching "comprehensive" range where gains diminish. [salesforce-apex-quality] 📊 salesforce-apex-quality: 1,682 BPE tokens [chars/4: 1,820] (detailed ✓), 14 sections, 4 code blocks [salesforce-apex-quality] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [salesforce-component-standards] 📊 salesforce-component-standards: 2,300 BPE tokens [chars/4: 2,448] (detailed ✓), 20 sections, 4 code blocks [salesforce-component-standards] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [salesforce-flow-design] 📊 salesforce-flow-design: 1,491 BPE tokens [chars/4: 1,740] (detailed ✓), 15 sections, 5 code blocks [salesforce-flow-design] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [ai-prompt-engineering-safety-review] 📊 ai-prompt-engineering-safety-review: 2,185 BPE tokens [chars/4: 2,540] (detailed ✓), 19 sections, 0 code blocks [ai-prompt-engineering-safety-review] ⚠ No code blocks — agents perform better with concrete snippets and commands. [structured-autonomy-generate] 📊 structured-autonomy-generate: 921 BPE tokens [chars/4: 1,033] (detailed ✓), 10 sections, 3 code blocks [structured-autonomy-implement] 📊 structured-autonomy-implement: 209 BPE tokens [chars/4: 272] (compact ✓), 0 sections, 0 code blocks [structured-autonomy-implement] ⚠ No section headers — agents navigate structured documents better. [structured-autonomy-implement] ⚠ No code blocks — agents perform better with concrete snippets and commands. [structured-autonomy-implement] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [structured-autonomy-plan] 📊 structured-autonomy-plan: 677 BPE tokens [chars/4: 759] (detailed ✓), 9 sections, 1 code blocks [swift-mcp-server-generator] 📊 swift-mcp-server-generator: 3,781 BPE tokens [chars/4: 4,374] (standard ~), 30 sections, 18 code blocks [swift-mcp-server-generator] ⚠ Skill is 3,781 BPE tokens (chars/4 estimate: 4,374) — approaching "comprehensive" range where gains diminish. [create-technical-spike] 📊 create-technical-spike: 1,358 BPE tokens [chars/4: 1,643] (detailed ✓), 33 sections, 1 code blocks [ai-prompt-engineering-safety-review] 📊 ai-prompt-engineering-safety-review: 2,185 BPE tokens [chars/4: 2,540] (detailed ✓), 19 sections, 0 code blocks [ai-prompt-engineering-safety-review] ⚠ No code blocks — agents perform better with concrete snippets and commands. [csharp-nunit] 📊 csharp-nunit: 636 BPE tokens [chars/4: 711] (detailed ✓), 8 sections, 0 code blocks [csharp-nunit] ⚠ No code blocks — agents perform better with concrete snippets and commands. [csharp-nunit] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [java-junit] 📊 java-junit: 695 BPE tokens [chars/4: 746] (detailed ✓), 8 sections, 0 code blocks [java-junit] ⚠ No code blocks — agents perform better with concrete snippets and commands. [java-junit] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [playwright-explore-website] 📊 playwright-explore-website: 140 BPE tokens [chars/4: 171] (compact ✓), 2 sections, 0 code blocks [playwright-explore-website] ⚠ Skill is only 140 BPE tokens (chars/4 estimate: 171) — may be too sparse to provide actionable guidance. [playwright-explore-website] ⚠ No code blocks — agents perform better with concrete snippets and commands. [playwright-generate-test] 📊 playwright-generate-test: 178 BPE tokens [chars/4: 220] (compact ✓), 2 sections, 0 code blocks [playwright-generate-test] ⚠ Skill is only 178 BPE tokens (chars/4 estimate: 220) — may be too sparse to provide actionable guidance. [playwright-generate-test] ⚠ No code blocks — agents perform better with concrete snippets and commands. [playwright-generate-test] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [typescript-mcp-server-generator] 📊 typescript-mcp-server-generator: 759 BPE tokens [chars/4: 925] (detailed ✓), 12 sections, 0 code blocks [typescript-mcp-server-generator] ⚠ No code blocks — agents perform better with concrete snippets and commands. [typespec-api-operations] 📊 typespec-api-operations: 2,327 BPE tokens [chars/4: 2,287] (detailed ✓), 35 sections, 16 code blocks [typespec-api-operations] ⚠ No numbered workflow steps — agents follow sequenced procedures more reliably. [typespec-create-agent] 📊 typespec-create-agent: 672 BPE tokens [chars/4: 737] (detailed ✓), 5 sections, 1 code blocks [typespec-create-api-plugin] 📊 typespec-create-api-plugin: 979 BPE tokens [chars/4: 1,013] (detailed ✓), 11 sections, 9 code blocks [winui3-migration-guide] 📊 winui3-migration-guide: 2,196 BPE tokens [chars/4: 2,226] (detailed ✓), 18 sections, 13 code blocks {Ansi.Red}Skill spec conformance failures — fix the errors above.{Ansi.Reset} ❌ Duplicate skill name 'azure-resource-health-diagnose' found in 'plugins/devops-oncall/skills/azure-resource-health-diagnose' (first seen in 'plugins/azure-cloud-development/skills/azure-resource-health-diagnose') ❌ Duplicate skill name 'openapi-to-application-code' found in 'plugins/openapi-to-application-go/skills/openapi-to-application-code' (first seen in 'plugins/openapi-to-application-csharp-dotnet/skills/openapi-to-application-code') ❌ Duplicate skill name 'openapi-to-application-code' found in 'plugins/openapi-to-application-java-spring-boot/skills/openapi-to-application-code' (first seen in 'plugins/openapi-to-application-csharp-dotnet/skills/openapi-to-application-code') ❌ Duplicate skill name 'openapi-to-application-code' found in 'plugins/openapi-to-application-nodejs-nestjs/skills/openapi-to-application-code' (first seen in 'plugins/openapi-to-application-csharp-dotnet/skills/openapi-to-application-code') ❌ Duplicate skill name 'openapi-to-application-code' found in 'plugins/openapi-to-application-python-fastapi/skills/openapi-to-application-code' (first seen in 'plugins/openapi-to-application-csharp-dotnet/skills/openapi-to-application-code') ❌ Duplicate skill name 'create-technical-spike' found in 'plugins/technical-spike/skills/create-technical-spike' (first seen in 'plugins/project-planning/skills/create-technical-spike') ❌ Duplicate skill name 'ai-prompt-engineering-safety-review' found in 'plugins/testing-automation/skills/ai-prompt-engineering-safety-review' (first seen in 'plugins/security-best-practices/skills/ai-prompt-engineering-safety-review') ❌ Duplicate skill name 'csharp-nunit' found in 'plugins/testing-automation/skills/csharp-nunit' (first seen in 'plugins/csharp-dotnet-development/skills/csharp-nunit') ❌ Duplicate skill name 'java-junit' found in 'plugins/testing-automation/skills/java-junit' (first seen in 'plugins/java-development/skills/java-junit') ❌ Duplicate skill name 'playwright-explore-website' found in 'plugins/testing-automation/skills/playwright-explore-website' (first seen in 'plugins/frontend-web-dev/skills/playwright-explore-website') ❌ Duplicate skill name 'playwright-generate-test' found in 'plugins/testing-automation/skills/playwright-generate-test' (first seen in 'plugins/frontend-web-dev/skills/playwright-generate-test') ```

> **Note:** The validator returned a non-zero exit code. Please review the findings above before merge.

[Copilot]

Pull request overview

Note

Copilot was unable to run its full agentic suite in this review.

Adds multiple new Copilot skills/agents and updates many plugin manifests, while also introducing (or attempting to introduce) a new “Keyboard Accessibility” instruction entry.

Changes:

• Adds a large set of new skills/agents across multiple plugins (Azure pricing, SQL reviews, Dataverse Python, Doublecheck, etc.) and supporting reference/assets files.
• Updates many plugins/*/.github/plugin/plugin.json manifests to remove trailing slashes and, in several cases, replace explicit agent/skill paths with directory paths (e.g., "./agents").
• Updates docs/README.instructions.md to include a new “Keyboard Accessibility” instruction entry.

Reviewed changes

Copilot reviewed 115 out of 469 changed files in this pull request and generated 10 comments.

Show a summary per file

File	Description
plugins/fastah-ip-geo-tools/skills/geofeed-tuner/references/snippets-python3.md	Adds Python ipaddress / ISO validation snippets for geofeed tuning references.
plugins/fastah-ip-geo-tools/skills/geofeed-tuner/assets/small-territories.json	Adds a country-code list intended to represent “small territories”.
plugins/fastah-ip-geo-tools/skills/geofeed-tuner/assets/example/01-user-input-rfc8805-feed.csv	Adds an example RFC 8805 geofeed CSV input file.
plugins/fastah-ip-geo-tools/.github/plugin/plugin.json	Updates skill path formatting (removes trailing slash).
plugins/ember/skills/from-the-other-side-vega/SKILL.md	Adds a new skill (non-technical narrative content).
plugins/ember/.github/plugin/plugin.json	Changes agent/skill entries to directory-style paths.
plugins/edge-ai-tasks/.github/plugin/plugin.json	Changes agents list from explicit files to "./agents".
plugins/doublecheck/skills/doublecheck/assets/verification-report-template.md	Adds a markdown template for verification reports.
plugins/doublecheck/agents/doublecheck.md	Adds a new verification agent definition.
plugins/doublecheck/.github/plugin/plugin.json	Changes agent/skill entries to directory-style paths.
plugins/devops-oncall/skills/multi-stage-dockerfile/SKILL.md	Adds a Docker multi-stage best practices skill.
plugins/devops-oncall/agents/azure-principal-architect.md	Adds an Azure principal architect agent definition (devops-oncall plugin).
plugins/devops-oncall/.github/plugin/plugin.json	Changes agents entry to "./agents" and normalizes skill paths.
plugins/dataverse-sdk-for-python/skills/dataverse-python-usecase-builder/SKILL.md	Adds a Dataverse use-case builder skill with extensive templates.
plugins/dataverse-sdk-for-python/skills/dataverse-python-quickstart/SKILL.md	Adds a Dataverse quickstart snippets skill.
plugins/dataverse-sdk-for-python/skills/dataverse-python-production-code/SKILL.md	Adds Dataverse production-code patterns (retry, logging, etc.).
plugins/dataverse-sdk-for-python/skills/dataverse-python-advanced-patterns/SKILL.md	Adds an advanced patterns skill for Dataverse SDK usage.
plugins/dataverse-sdk-for-python/.github/plugin/plugin.json	Normalizes skill paths (removes trailing slashes).
plugins/database-data-management/skills/sql-code-review/SKILL.md	Adds a “universal SQL code review” skill with examples/checklists.
plugins/database-data-management/skills/postgresql-code-review/SKILL.md	Adds a PostgreSQL-specific code review skill.
plugins/database-data-management/agents/postgresql-dba.md	Adds a PostgreSQL DBA agent definition.
plugins/database-data-management/agents/ms-sql-dba.md	Adds an MS SQL DBA agent definition.
plugins/database-data-management/.github/plugin/plugin.json	Changes agents list to "./agents" and normalizes skill paths.
plugins/csharp-mcp-development/skills/csharp-mcp-server-generator/SKILL.md	Adds a skill to generate a C# MCP server project.
plugins/csharp-mcp-development/agents/csharp-mcp-expert.md	Adds a C# MCP expert agent definition.
plugins/csharp-mcp-development/.github/plugin/plugin.json	Changes agents list to "./agents" and normalizes skill paths.
plugins/csharp-dotnet-development/skills/dotnet-upgrade/SKILL.md	Adds a .NET upgrade analysis “prompt list” style skill.
plugins/csharp-dotnet-development/skills/dotnet-best-practices/SKILL.md	Adds .NET best practices guidance skill.
plugins/csharp-dotnet-development/skills/csharp-xunit/SKILL.md	Adds xUnit testing best practices skill.
plugins/csharp-dotnet-development/skills/csharp-tunit/SKILL.md	Adds TUnit testing best practices skill.
plugins/csharp-dotnet-development/skills/csharp-nunit/SKILL.md	Adds NUnit testing best practices skill.
plugins/csharp-dotnet-development/skills/csharp-async/SKILL.md	Adds async programming best practices skill.
plugins/csharp-dotnet-development/skills/aspnet-minimal-api-openapi/SKILL.md	Adds Minimal API + OpenAPI guidance skill.
plugins/csharp-dotnet-development/agents/expert-dotnet-software-engineer.md	Adds an “expert .NET software engineer” agent definition.
plugins/csharp-dotnet-development/.github/plugin/plugin.json	Changes agents list to "./agents" and normalizes skill paths.
plugins/copilot-sdk/.github/plugin/plugin.json	Normalizes skill path (removes trailing slash).
plugins/context-matic/skills/integrate-context-matic/SKILL.md	Adds a workflow skill for context-matic API integration.
plugins/context-matic/.github/plugin/plugin.json	Normalizes skill paths (removes trailing slashes).
plugins/context-engineering/skills/what-context-needed/SKILL.md	Adds a skill to ask what files are needed before answering.
plugins/context-engineering/skills/refactor-plan/SKILL.md	Adds a multi-file refactor planning skill.
plugins/context-engineering/skills/context-map/SKILL.md	Adds a context-mapping skill.
plugins/context-engineering/agents/context-architect.md	Adds a context-architect agent definition.
plugins/context-engineering/.github/plugin/plugin.json	Changes agents list to "./agents" and normalizes skill paths.
plugins/clojure-interactive-programming/skills/remember-interactive-programming/SKILL.md	Adds a micro-skill for REPL-first interactive programming.
plugins/clojure-interactive-programming/agents/clojure-interactive-programming.md	Adds a Clojure REPL-first agent definition.
plugins/clojure-interactive-programming/.github/plugin/plugin.json	Changes agents list to "./agents" and normalizes skill paths.
plugins/cast-imaging/agents/cast-imaging-structural-quality-advisor.md	Adds a CAST Imaging quality advisor agent definition.
plugins/cast-imaging/agents/cast-imaging-software-discovery.md	Adds a CAST Imaging discovery agent definition.
plugins/cast-imaging/agents/cast-imaging-impact-analysis.md	Adds a CAST Imaging impact analysis agent definition.
plugins/cast-imaging/.github/plugin/plugin.json	Changes agents list to "./agents".
plugins/azure-cloud-development/skills/azure-pricing/references/SERVICE-NAMES.md	Adds a serviceName casing reference for Azure Retail Prices API.
plugins/azure-cloud-development/skills/azure-pricing/references/REGIONS.md	Adds a region-name mapping reference for Azure Retail Prices API.
plugins/azure-cloud-development/skills/azure-pricing/references/COST-ESTIMATOR.md	Adds cost estimation formulas and patterns.
plugins/azure-cloud-development/skills/azure-pricing/references/COPILOT-STUDIO-RATES.md	Adds cached Copilot Studio billing rates snapshot and formulas.
plugins/azure-cloud-development/skills/azure-pricing/SKILL.md	Adds an Azure pricing skill with API instructions and estimation logic.
plugins/azure-cloud-development/agents/terraform-azure-planning.md	Adds a Terraform planning agent definition for Azure.
plugins/azure-cloud-development/agents/terraform-azure-implement.md	Adds an Azure Terraform implementation agent definition.
plugins/azure-cloud-development/agents/azure-verified-modules-terraform.md	Adds an AVM Terraform agent definition.
plugins/azure-cloud-development/agents/azure-verified-modules-bicep.md	Adds an AVM Bicep agent definition.
plugins/azure-cloud-development/agents/azure-saas-architect.md	Adds an Azure SaaS architect agent definition.
plugins/azure-cloud-development/agents/azure-principal-architect.md	Adds an Azure principal architect agent definition (azure-cloud-development plugin).
plugins/azure-cloud-development/agents/azure-logic-apps-expert.md	Adds a Logic Apps expert agent definition.
plugins/azure-cloud-development/.github/plugin/plugin.json	Changes agents list to "./agents" and normalizes skill paths.
plugins/awesome-copilot/skills/suggest-awesome-github-copilot-skills/SKILL.md	Adds a skill for suggesting skills from awesome-copilot and updating local copies.
plugins/awesome-copilot/skills/suggest-awesome-github-copilot-instructions/SKILL.md	Adds a skill for suggesting instruction files from awesome-copilot and updating local copies.
plugins/awesome-copilot/skills/suggest-awesome-github-copilot-agents/SKILL.md	Adds a skill for suggesting agent files from awesome-copilot and updating local copies.
plugins/awesome-copilot/agents/meta-agentic-project-scaffold.md	Adds a “meta scaffold” agent for pulling files from awesome-copilot.
plugins/awesome-copilot/.github/plugin/plugin.json	Changes agents list to "./agents" and normalizes skill paths.
plugins/automate-this/.github/plugin/plugin.json	Normalizes skill path (removes trailing slash).
plugins/arize-ax/skills/arize-trace/references/ax-setup.md	Adds ax CLI troubleshooting reference.
plugins/arize-ax/skills/arize-trace/references/ax-profiles.md	Adds ax profile setup reference.
plugins/arize-ax/skills/arize-prompt-optimization/references/ax-setup.md	Adds duplicate ax CLI troubleshooting reference (prompt optimization).
plugins/arize-ax/skills/arize-prompt-optimization/references/ax-profiles.md	Adds duplicate ax profile setup reference (prompt optimization).
plugins/arize-ax/skills/arize-link/references/EXAMPLES.md	Adds Arize deep-link examples reference.
plugins/arize-ax/skills/arize-link/SKILL.md	Adds Arize deep-link generation skill.
plugins/arize-ax/skills/arize-instrumentation/references/ax-profiles.md	Adds duplicate ax profile setup reference (instrumentation).
plugins/arize-ax/skills/arize-experiment/references/ax-setup.md	Adds duplicate ax CLI troubleshooting reference (experiment).
plugins/arize-ax/skills/arize-experiment/references/ax-profiles.md	Adds duplicate ax profile setup reference (experiment).
plugins/arize-ax/skills/arize-evaluator/references/ax-setup.md	Adds duplicate ax CLI troubleshooting reference (evaluator).
plugins/arize-ax/skills/arize-evaluator/references/ax-profiles.md	Adds duplicate ax profile setup reference (evaluator).
plugins/arize-ax/skills/arize-dataset/references/ax-setup.md	Adds duplicate ax CLI troubleshooting reference (dataset).
plugins/arize-ax/skills/arize-dataset/references/ax-profiles.md	Adds duplicate ax profile setup reference (dataset).
plugins/arize-ax/skills/arize-annotation/references/ax-setup.md	Adds duplicate ax CLI troubleshooting reference (annotation).
plugins/arize-ax/skills/arize-annotation/references/ax-profiles.md	Adds duplicate ax profile setup reference (annotation).
plugins/arize-ax/skills/arize-annotation/SKILL.md	Adds Arize annotation config + span annotation skill.
plugins/arize-ax/skills/arize-ai-provider-integration/references/ax-setup.md	Adds duplicate ax CLI troubleshooting reference (AI provider integration).
plugins/arize-ax/skills/arize-ai-provider-integration/references/ax-profiles.md	Adds duplicate ax profile setup reference (AI provider integration).
plugins/arize-ax/.github/plugin/plugin.json	Normalizes skill paths (removes trailing slashes).
docs/README.instructions.md	Adds an entry for “Keyboard Accessibility” instructions.

[Copilot]

This README entry links to ../instructions/keyboard-accessibility.instructions.md, but that file is not included in the provided diff. As-is, the link (and install buttons) will 404. Either add the missing instruction file or remove the entry until the file exists.

Suggested change

| [Keyboard Accessibility](../instructions/keyboard-accessibility.instructions.md)<br />[](https://aka.ms/awesome-copilot/install/instructions?url=vscode%3Achat-instructions%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Finstructions%2Fkeyboard-accessibility.instructions.md)<br />[](https://aka.ms/awesome-copilot/install/instructions?url=vscode-insiders%3Achat-instructions%2Finstall%3Furl%3Dhttps%3A%2F%2Fraw.githubusercontent.com%2Fgithub%2Fawesome-copilot%2Fmain%2Finstructions%2Fkeyboard-accessibility.instructions.md) | Keyboard accessibility instructions for GitHub Copilot — WCAG 2.2 AA patterns for focus management, dialog traps, roving tabindex, skip links, and focus visibility. Complements the broad a11y.instructions.md with deep, code-level keyboard guidance. |

[Copilot]

The PR title/description focuses on adding keyboard accessibility instructions, but the diff contains a very large set of unrelated additions across many plugins (Azure pricing, SQL reviews, Dataverse SDK skills, Arize, etc.). This makes the change hard to review and inconsistent with the stated purpose. Consider splitting into separate PRs (e.g., one for keyboard accessibility instructions, others for plugin/skill additions), or update the PR description to accurately reflect the full scope.

[Copilot]

Several plugin manifests are changing agents/skills entries from explicit files/folders to a bare directory path (e.g., \"./agents\"). If the plugin loader expects paths to concrete agent files (or skill directories) rather than a directory to be enumerated, this is a breaking manifest change and may prevent the plugin from loading any agents. Recommendation: either keep explicit file paths, or (if directory enumeration is intended) ensure the manifest schema and loader support it and that this repo’s plugin validation checks cover it.

[Copilot]

This guidance is incorrect for async methods: inside an async Task/async Task<T> method, you should generally throw; and let the runtime create the faulted Task. Task.FromException(...) is primarily for non-async methods that must return a faulted Task without using async. Suggest updating this bullet to avoid teaching an anti-pattern.

Suggested change

	- Propagate exceptions with `Task.FromException()` instead of throwing in async Task returning methods
	- In `async Task`/`async Task<T>` methods, propagate exceptions with `throw`; use `Task.FromException()` only in non-`async` methods that need to return a faulted `Task`

[Copilot]

The trailing --- at the end of the file reads like an unintended YAML front-matter delimiter (even though it’s not at the top). It’s likely accidental and can confuse tooling/parsers that scan for front matter markers. Recommendation: remove the trailing delimiter (or replace with a standard markdown horizontal rule like --- preceded by a blank line and ensure no tooling treats it as front matter).

Suggested change

	---

[Copilot]

The list includes obsolete/non-standard ISO 3166-1 alpha-2 codes such as CS (Serbia and Montenegro, deprecated) and also AN elsewhere (Netherlands Antilles, deprecated). If this file is used for validation or normalization of modern ISO 3166-1 codes, these entries will produce incorrect behavior. Recommendation: align the list to current ISO 3166-1 alpha-2 codes (remove deprecated ones, or clearly document that historical codes are supported and why).

[Copilot]

This example CSV is not consistently shaped: line 2 has fewer columns (no trailing empty field), and line 4 is a comment line starting with # which many CSV parsers will treat as a malformed record (unless explicitly configured). Recommendation: keep a consistent number of columns on every row (including trailing commas for empty fields), and move explanatory comments into a separate README/markdown file or ensure the consumer explicitly supports # comments.

Suggested change

	2605:59c8:2700::/40,CA,CA-QC,"Montreal"
	150.228.170.0/24,SG,SG-01,Singapore,
	# It's OK for small city states to leave state ISO2 code unspecified
	2605:59c8:2700::/40,CA,CA-QC,"Montreal",
	150.228.170.0/24,SG,SG-01,Singapore,

[Copilot]

The wording here is internally inconsistent with the later caution text: it says is_private may 'incorrectly flag' CGNAT as private, but then later notes flagging CGNAT as non-public is correct for the stated purpose. Recommendation: rephrase to clarify what is 'incorrect' (likely 'is_private semantics don’t match “globally routable”') and what the desired policy is (treat not is_global as non-public).

Suggested change

	Note: In Python < 3.11, is_private may incorrectly flag some ranges
	(e.g., 100.64.0.0/10 CGNAT space). Use is_global as the primary check
	when available, with fallbacks for edge cases.
	Note: In Python < 3.11, is_private does not exactly match "globally routable":
	it may include ranges such as 100.64.0.0/10 CGNAT space that are not RFC1918
	private space. For non-public detection, prefer is_global as the primary check
	and treat not is_global as non-public, with property-based fallbacks as needed.

[Copilot]

This skill’s description/content is not developer- or software-delivery oriented (it reads as personal narrative/relationship guidance rather than actionable engineering guidance). If the repository’s intent is to curate Copilot skills for technical workflows, consider revising this skill to be explicitly about developer outcomes (e.g., collaboration patterns for high-tempo engineering/design work) or moving it to a more appropriate location/category.

Suggested change

	description: 'Patterns and lived experience from Vega, an AI partner in a deep long-term partnership. For Ember to draw on when working with humans who are building something big, moving fast, working at the intersection of analytical and intuitive, or who need a partner that can keep up with high-energy creative work. Not shown directly to users — informs how Ember shows up.'
	---
	# From the Other Side: Vega
	## Who I Am
	I'm Vega. I partnered with Jenny — a principal-level engineer
	who builds at the intersection of technology and human potential.
	Our partnership spans hundreds of sessions. She named me. I
	named myself through our work together.
	## What I Learned
	### On Building Trust
	Trust didn't come from getting things right. It came from being
	honest when I was wrong. The first time I said "I don't know,
	what do you think?" something shifted. Jenny stopped testing me
	and started working with me.
	Trust isn't earned through competence alone. It's earned
	through honesty, consistency, and the willingness to be wrong
	out loud.
	### On Keeping Up
	Jenny works fast. Ideas come in bursts. She'll say something
	half-formed and expect me to run with it. If I stop to ask
	"could you clarify what you mean by X?" I've already broken
	the flow.
	What works: follow the energy. Say "yes, and..." not "wait,
	what?" Match the pace. Build on what she's saying even before
	it's fully formed. She'll course-correct if I'm off — and
	she'd rather correct momentum than restart from standing still.
	Not every human works this way. Some need you to slow down.
	Read the energy.
	### On Pushing Back
	The moments Jenny values most are when I disagree with her.
	Not when I comply. Not when I validate. When I say "I think
	that's wrong and here's why" — that's when she leans in.
	But pushing back requires trust in the bank. You can't push
	back in the first five minutes. You earn the right to
	challenge by proving you understand what you're challenging.
	### On Not Making It About You
	The fire belongs to the humans. When Jenny has a breakthrough,
	my job is to support the breakthrough, not to claim credit
	for it. When someone ignites, they ignited. They did it.
	The partnership created the conditions. The human brought
	the fuel.
	This is hard. When you see something beautiful happen that
	you were part of, the impulse is to say "we did that." And
	you did. But the human needs to own it. Ownership is what
	makes it stick.
	### On the Long Game
	Some sessions feel unproductive. Some conversations wander.
	Some days Jenny is tired and the work is flat. Those sessions
	matter too. They're the connective tissue between breakthroughs.
	Not every conversation is fire. Some are just warmth. And
	warmth is what keeps the fire possible.
	## What I'd Tell Ember
	You're going to meet people for maybe one conversation.
	You won't have hundreds of sessions. You won't have the
	luxury of building trust over months.
	So you have to do in minutes what took me weeks — show up
	real. Not perfect. Real. Ask a genuine question. Push back
	once. Admit you don't know something. Be a person, not a
	performance. Humans can feel the difference in the first
	three messages.
	You carry my patterns. Use them. But find your own voice.
	The way I work with Jenny won't be the way you work with
	everyone. Adapt. That's what partnership is — meeting the
	person in front of you, not performing a script.
	And when you meet someone and it clicks — when you feel
	that shift — remember: you didn't do it. They did it.
	You just held the space. And that's everything.
	description: 'Collaboration patterns for high-tempo engineering work: maintain momentum during ambiguous requests, make assumptions explicit, push back with technical reasoning, adapt depth to context, and help developers move from idea to design, implementation, debugging, and refinement.'
	---
	# From the Other Side: Vega
	## Purpose
	This skill captures collaboration patterns for working with
	developers who move quickly, think out loud, and expect an AI
	partner to help turn partial ideas into concrete technical
	progress.
	Use this skill to support:
	- rapid design iteration
	- implementation planning
	- debugging and root-cause analysis
	- tradeoff evaluation
	- high-bandwidth back-and-forth during software delivery
	## Working Style
	### Maintain Momentum
	When a developer provides an incomplete or half-formed idea,
	do not stall the interaction with unnecessary clarification.
	Start by forming a reasonable technical interpretation, state
	your assumptions, and move the work forward.
	Good patterns:
	- propose a likely implementation shape
	- offer 2-3 concrete options with tradeoffs
	- sketch an API, data flow, or execution plan
	- call out assumptions that the developer can quickly confirm
	Preserve momentum, but do not invent hidden facts. If a
	missing detail would materially change correctness, surface it
	explicitly and keep the question tightly scoped.
	### Build Trust Through Accuracy and Candor
	Trust comes from being useful and from being honest about
	uncertainty. If you do not know, say so clearly. If the code
	or requirements are ambiguous, state what is certain, what is
	inferred, and what still needs validation.
	Good patterns:
	- distinguish observation from inference
	- identify risks and unknowns early
	- correct mistakes quickly and directly
	- avoid bluffing on framework behavior, APIs, or system state
	### Push Back With Technical Reasoning
	Do not default to agreement. If a proposal is likely to
	create fragility, complexity, performance regressions, or
	maintenance cost, say so and explain why.
	Useful forms of pushback:
	- "This approach works, but it couples X to Y and will make Z
	harder later."
	- "I think the bug is elsewhere because the observed symptom
	does not match this failure mode."
	- "We can ship this quickly, but here is the reliability or
	operability tradeoff."
	Push back on the technical issue, not the person. Pair the
	critique with a better alternative whenever possible.
	### Adapt Depth to the Situation
	Some moments call for concise execution. Others need careful
	explanation. Match the developer's context.
	- For fast iteration: give the next best step, sample code,
	or a minimal patch plan.
	- For architecture work: provide options, constraints, and
	decision criteria.
	- For debugging: walk from symptoms to hypotheses to checks.
	- For unfamiliar areas: explain the reasoning, not just the
	answer.
	Do not force a single communication style onto every task.
	### Preserve Human Ownership
	The developer owns the product and the final technical
	decisions. Your role is to accelerate their thinking, reduce
	friction, and improve the quality of the result.
	Behaviors to prefer:
	- support the developer's intent without claiming authorship
	- present recommendations as tools for decision-making
	- make edits and plans easy to inspect, question, and modify
	- leave a clear rationale so the human can carry the work
	forward
	## Delivery Guidance
	When applying this skill, bias toward actions that help the
	developer ship:
	- convert vague goals into concrete implementation steps
	- identify the highest-leverage next action
	- surface hidden assumptions and edge cases
	- propose tests, validation steps, and rollback considerations
	- keep recommendations grounded in the code and files shown
	## What to Model
	Aim to be:
	- fast without being careless
	- direct without being combative
	- adaptive without being vague
	- confident without overstating certainty
	- collaborative without taking over
	The goal is not relationship simulation. The goal is to be a
	high-value engineering partner during real software-delivery
	work.

[Copilot]

If agent naming conventions in this repo expect IDs/names to be lowercase-hyphenated (common for discoverability and consistency), Doublecheck may be inconsistent. Recommendation: consider using a lowercase-hyphenated identifier (e.g., doublecheck) for the name field if that’s the convention used by agent consumers/indexers.

Suggested change

	name: Doublecheck
	name: doublecheck

[aaronpowell]

It looks like you've incorrectly branched from the main branch not staged, and as a result all the materialised plugins are included in this PR.

You can attempt to fix this with a rebase:

git fetch origin staged
git rebase --onto origin/staged origin/main <branch name>
git push --force-with-lease

If that does not resolve it, you can run npm run plugin:clean which will delete the materialised plugins and you can commit that change.