Skip to content

Enable Datadog Claude Marketplace plugins#8218

Merged
NachoEchevarria merged 3 commits intomasterfrom
nacho/UseDDSkills
Feb 19, 2026
Merged

Enable Datadog Claude Marketplace plugins#8218
NachoEchevarria merged 3 commits intomasterfrom
nacho/UseDDSkills

Conversation

@NachoEchevarria
Copy link
Collaborator

@NachoEchevarria NachoEchevarria commented Feb 18, 2026 •
edited
Loading

Summary of changes

Adds .claude/settings.json to enable the Datadog Claude Marketplace plugins, with irrelevant skills filtered out via permission deny rules.

No manual setup required — the settings.json handles marketplace registration and plugin activation automatically when Claude Code is launched from this repo.

Reason for change

Gives contributors access to shared Datadog-wide Claude Code tooling (CI debugging, PR workflows, employee lookup, Jira integration) that complements the repo-specific local skills already defined under .claude/skills/.

Implementation details

Enables the dd plugin bundle from the marketplace, then uses permissions.deny rules to block skills not relevant to dd-trace-dotnet.

Marketplace skills — allowed

Skill Purpose
dd:\ci:fix Analyze CI failures and propose fixes
dd:fetch-ci-results Fetch CI logs, detect flakes, retry jobs
dd:\pr:address-feedback Address PR review feedback
dd:get-pr-feedback Fetch and structure PR review comments
dd:jira-ticket-solver Jira ticket workflow automation
dd:whoisthis Employee lookup by email/GitHub username

Marketplace skills — denied

Skill Reason
dd:conductor Deployment platform — not used by dd-trace-dotnet
dd:trebuchet logs-backend repo only
dd:render-helm Helm charts for dd-source services
dd:deploy-yaml-local-dev deploy.yaml for dd-source services
dd:integrate devflow integration branches — not used in this repo
dd:live-test-api Live Datadog API calls — not needed for tracer development
dd:debug-ddci-request DDCI request debugging — covered by local CI skills

Existing local skills (unchanged)

Skill Purpose
/analyze-crash Native crash stack trace analysis
/analyze-error Managed error analysis with fix recommendations
/review-pr PR review with inline GitHub comments
/troubleshoot-ci-build Azure DevOps pipeline troubleshooting

Test coverage

Configuration-only change, no code changes.

Other details

The DataDog/claude-marketplace repo is private. For external contributors, the marketplace plugins will simply fail to load — local skills and core Claude Code functionality remain fully available.

@NachoEchevarria NachoEchevarria requested a review from a team as a code owner February 18, 2026 17:52
@NachoEchevarria NachoEchevarria added the AI Generated Largely based on code generated by an AI or LLM. This label is the same across all dd-trace-* repos label Feb 18, 2026
@pr-commenter
Copy link

pr-commenter bot commented Feb 18, 2026 •
edited
Loading

Benchmarks

Benchmark execution time: 2026-02-19 12:57:04

Comparing candidate commit 9938c6b in PR branch nacho/UseDDSkills with baseline commit e0e1029 in branch master.

Found 8 performance improvements and 6 performance regressions! Performance is the same for 163 metrics, 15 unstable metrics.

scenario:Benchmarks.Trace.AgentWriterBenchmark.WriteAndFlushEnrichedTraces net6.0

  • 🟩 execution_time [-98.974ms; -98.895ms] or [-49.384%; -49.344%]

scenario:Benchmarks.Trace.Asm.AppSecBodyBenchmark.ObjectExtractorMoreComplexBody net6.0

  • 🟩 execution_time [-17.519ms; -11.762ms] or [-8.258%; -5.544%]

scenario:Benchmarks.Trace.Asm.AppSecEncoderBenchmark.EncodeLegacyArgs net6.0

  • 🟥 execution_time [+24.006ms; +24.164ms] or [+13.411%; +13.500%]

scenario:Benchmarks.Trace.Asm.AppSecEncoderBenchmark.EncodeLegacyArgs netcoreapp3.1

  • 🟩 execution_time [-20.111ms; -19.090ms] or [-10.063%; -9.552%]

scenario:Benchmarks.Trace.AspNetCoreBenchmark.SendRequest net6.0

  • 🟩 execution_time [-29.034ms; -28.279ms] or [-24.059%; -23.433%]

scenario:Benchmarks.Trace.AspNetCoreBenchmark.SendRequest netcoreapp3.1

  • 🟥 throughput [-1385.543op/s; -580.347op/s] or [-13.447%; -5.632%]

scenario:Benchmarks.Trace.CIVisibilityProtocolWriterBenchmark.WriteAndFlushEnrichedTraces net6.0

  • 🟥 execution_time [+35.770ms; +40.650ms] or [+20.369%; +23.148%]

scenario:Benchmarks.Trace.CIVisibilityProtocolWriterBenchmark.WriteAndFlushEnrichedTraces netcoreapp3.1

  • 🟥 throughput [-474.824op/s; -364.695op/s] or [-27.626%; -21.219%]

scenario:Benchmarks.Trace.CharSliceBenchmark.OptimizedCharSliceWithPool netcoreapp3.1

  • 🟥 execution_time [+110.450µs; +125.190µs] or [+5.924%; +6.715%]
  • 🟥 throughput [-33.833op/s; -29.918op/s] or [-6.307%; -5.578%]

scenario:Benchmarks.Trace.DbCommandBenchmark.ExecuteNonQuery netcoreapp3.1

  • 🟩 execution_time [-14.938ms; -11.667ms] or [-7.197%; -5.621%]

scenario:Benchmarks.Trace.GraphQLBenchmark.ExecuteAsync net472

  • 🟩 throughput [+22350.663op/s; +23491.521op/s] or [+5.733%; +6.026%]

scenario:Benchmarks.Trace.Iast.StringAspectsBenchmark.StringConcatBenchmark netcoreapp3.1

  • 🟩 throughput [+2203.440op/s; +3890.361op/s] or [+12.466%; +22.010%]

scenario:Benchmarks.Trace.SingleSpanAspNetCoreBenchmark.SingleSpanAspNetCore netcoreapp3.1

  • 🟩 throughput [+14971710.461op/s; +16055212.274op/s] or [+6.650%; +7.131%]

pierotibou
pierotibou approved these changes Feb 18, 2026
View reviewed changes
Copy link
Contributor

@pierotibou pierotibou left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I tried the coach me skill earlier today, it was interesting to see i wasn't doing some basic stuff.

andrewlock
andrewlock reviewed Feb 19, 2026
View reviewed changes
Copy link
Member

@andrewlock andrewlock left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A bunch of these aren't applicable to us, and could cause confusion. Is there a way to specify which specific plugins to enable? 🤔

@NachoEchevarria @claude
Filter marketplace skills to those relevant for dd-trace-dotnet
1211d44 
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@NachoEchevarria
Copy link
Collaborator Author

NachoEchevarria commented Feb 19, 2026

A bunch of these aren't applicable to us, and could cause confusion. Is there a way to specify which specific plugins to enable? 🤔

Good point. I initially thought that it was not possible, but there is a way. I have updated the plugins, eliminating the ones that we dont need. Thanks!

@NachoEchevarria @claude
Deny dd:debug-ddci-request skill
9938c6b 
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@NachoEchevarria
Copy link
Collaborator Author

NachoEchevarria commented Feb 19, 2026

A bunch of these aren't applicable to us, and could cause confusion. Is there a way to specify which specific plugins to enable? 🤔

Good point. I initially thought that it was not possible, but there is a way. I have updated the plugins, eliminating the ones that we dont need. Thanks!

BTW, this approach denies using these skills but they are listed if run the /skill command even though if you try to use them, you can't but AFAIK there is no way to unlist them without removing all the skills

andrewlock
andrewlock approved these changes Feb 19, 2026
View reviewed changes
Copy link
Member

@andrewlock andrewlock left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks!

Do dd:\ci:fix and dd:fetch-ci-results actually work? I bet they assume GitLab, right?

@NachoEchevarria
Copy link
Collaborator Author

NachoEchevarria commented Feb 19, 2026

Thanks!

Do dd:\ci:fix and dd:fetch-ci-results actually work? I bet they assume GitLab, right?

Yes, that's gitlab and GH workflows. I have tested it and it works. Actually, I was thinking about adding a new feature in a subsequent PR because now we have two skills: one for gitlab/GH and another for azure. While both are different domains that require different approaches, we should proably unify them in a single CI analyzer skill that can call any of them to analyze a full .net tracer pipeline.

@NachoEchevarria
Copy link
Collaborator Author

NachoEchevarria commented Feb 19, 2026

Thanks for your feedback and reviews!

@NachoEchevarria NachoEchevarria merged commit 8941a2d into master Feb 19, 2026
98 of 101 checks passed
@NachoEchevarria NachoEchevarria deleted the nacho/UseDDSkills branch February 19, 2026 13:41
@github-actions github-actions bot added this to the vNext-v3 milestone Feb 19, 2026
@andrewlock andrewlock added the area:builds project files, build scripts, pipelines, versioning, releases, packages label Feb 23, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@andrewlock andrewlock andrewlock approved these changes

@pierotibou pierotibou pierotibou approved these changes

Assignees

No one assigned

Labels

AI Generated Largely based on code generated by an AI or LLM. This label is the same across all dd-trace-* repos area:builds project files, build scripts, pipelines, versioning, releases, packages

Projects

None yet

Milestone

3.38.0

Development

Successfully merging this pull request may close these issues.

3 participants

@NachoEchevarria @andrewlock @pierotibou