Skip to content

docs(agents): simplify repository guidance - #9969

Merged
bengl merged 14 commits into
masterfrom
pabloerhard/refactor-agents-md
Aug 28, 2026
Merged

bengl merged 14 commits into
masterfrom
pabloerhard/refactor-agents-md

Conversation

@pabloerhard

@pabloerhard pabloerhard commented Aug 25, 2026 •

Copy link
Copy Markdown
Contributor

What does this PR do?

Condenses the root AGENTS.md from a broad engineering handbook into concise repository-wide setup, workflow, testing, safety, backportability, and CI guidance.

Moves detailed architecture-review and flaky-test investigation procedures into on-demand project skills, exposes the canonical .agents/skills catalog to Claude and Cursor through symlinks, and assigns product-specific CODEOWNERS for those skills.

Motivation

The root instructions had grown large enough that every agent task consumed detailed guidance for specialized workflows regardless of relevance. Keeping universal invariants in AGENTS.md and loading task-specific skills on demand reduces context usage while preserving important engineering policies.

This structure also avoids premature nested AGENTS.md files when the specialized guidance is task-shaped rather than directory-shaped.

Additional Notes

  • Adds architecture-review with the six-dimension baseline-to-proposal review framework.
  • Adds flaky-test-investigation with evidence-based classification and optional parallel sub-agent investigation.
  • Keeps .agents/skills as the canonical source; .claude/skills and .cursor/skills entries are symlinks.
  • Assigns architecture review to SDK Capabilities, flaky-test investigation to Language Platform, LLMObs skills to ML Observability, and serverless integrations to APM Serverless across all skill catalogs.
  • Validation: git diff --check, skill metadata and line-length checks, newline checks, and symlink/catalog integrity checks.
  • npm run lint:codeowners:ci was attempted but could not run because codeowners-audit is not installed in this worktree.
  • No code tests were run because this PR only changes agent instructions, skill metadata, symlinks, and CODEOWNERS.
  • The local pre-commit hook was skipped by Git because .husky/pre-commit is not executable.

@dd-octo-sts

dd-octo-sts Bot commented Aug 25, 2026 •

Copy link
Copy Markdown
Contributor

Overall package size

Self size: 8.56 MB
Deduped: 9.23 MB
No deduping: 9.23 MB

Dependency sizes | name | version | self size | total size | |------|---------|-----------|------------| | import-in-the-middle | 3.3.3 | 125.43 kB | 445.14 kB | | opentracing | 0.14.7 | 194.81 kB | 194.81 kB | | dc-polyfill | 0.1.11 | 25.74 kB | 25.74 kB |

🤖 This report was automatically generated by heaviest-objects-in-the-universe

@datadog-prod-us1-6

datadog-prod-us1-6 Bot commented Aug 25, 2026 •

Copy link
Copy Markdown

Tests

✅ All CI checks and tests passed.

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 98.59% (+0.01%)

This comment will be updated automatically if new data arrives.
🔗 Commit SHA: fb8b3a8 | Docs | View more details | Give us feedback!

@pr-commenter

pr-commenter Bot commented Aug 25, 2026 •

Copy link
Copy Markdown

Benchmarks

Benchmark execution time: 2026-08-28 21:43:24

Comparing candidate commit fb8b3a8 in PR branch pabloerhard/refactor-agents-md with baseline commit 20a55e7 in branch master.

📊 Benchmarking dashboard

Found 0 performance improvements and 0 performance regressions! Performance is the same for 2301 metrics, 9 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

  • 🟩 = significantly better candidate vs. baseline
  • 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

Unstable benchmarks

These benchmarks have a confidence interval too wide to call a change; treat them as noise rather than signal.

scenario:appsec-appsec-enabled-20

  • unstable max_rss_usage [-28.062MB; +4.810MB] or [-15.426%; +2.644%]

scenario:appsec-iast-with-vulnerability-iast-enabled-default-config-20

  • unstable max_rss_usage [-24708.953KB; +25191.353KB] or [-6.611%; +6.740%]

scenario:debugger-line-probe-with-snapshot-minimal-24

  • unstable max_rss_usage [-13844.839KB; +15783.239KB] or [-5.684%; +6.480%]

scenario:dogstatsd-with-tags-20

  • unstable cpu_user_time [-300.420ms; +385.401ms] or [-6.227%; +7.988%]
  • unstable execution_time [-305.265ms; +381.557ms] or [-6.226%; +7.782%]
  • unstable throughput [-135184.604op/s; +109410.577op/s] or [-7.895%; +6.390%]

scenario:plugin-claude-agent-sdk-compact-stream-scan-26

  • unstable cpu_usage_percentage [-5.579%; +5.132%]

scenario:plugin-graphql-long-with-depth-and-collapse-off-20

  • unstable max_rss_usage [-28.185MB; +35.557MB] or [-7.037%; +8.878%]

scenario:test-optimization-large-suite-20

  • unstable max_rss_usage [-5301.122KB; +7092.789KB] or [-6.493%; +8.687%]

@codecov

codecov Bot commented Aug 25, 2026 •

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.59%. Comparing base (20a55e7) to head (fb8b3a8).
⚠️ Report is 17 commits behind head on master.

Additional details and impacted files
@@            Coverage Diff             @@
##           master    #9969      +/-   ##
==========================================
+ Coverage   98.58%   98.59%   +0.01%     
==========================================
  Files         990      992       +2     
  Lines      149247   149804     +557     
  Branches    12763    12600     -163     
==========================================
+ Hits       147131   147697     +566     
+ Misses       2116     2107       -9     
Flag Coverage Δ
aiguard 63.39% <ø> (+1.13%) ⬆️
aiguard-integration 59.15% <ø> (+0.24%) ⬆️
apm-bucket-0 62.74% <ø> (+1.04%) ⬆️
apm-bucket-1 68.24% <ø> (+1.00%) ⬆️
apm-bucket-2 67.82% <ø> (+1.02%) ⬆️
apm-bucket-3 64.61% <ø> (+1.11%) ⬆️
apm-capabilities-tracing 62.60% <ø> (+0.22%) ⬆️
apm-integrations-aerospike 60.72% <ø> (+1.15%) ⬆️
apm-integrations-confluentinc-kafka-javascript 66.94% <ø> (+1.20%) ⬆️
apm-integrations-couchbase 61.24% <ø> (+1.16%) ⬆️
apm-integrations-http 66.49% <ø> (+0.98%) ⬆️
apm-integrations-kafkajs 67.53% <ø> (+1.15%) ⬆️
apm-integrations-next 64.09% <ø> (+1.07%) ⬆️
apm-integrations-prisma 62.09% <ø> (+0.92%) ⬆️
appsec 77.12% <ø> (+0.88%) ⬆️
appsec-express_fastify_graphql 73.53% <ø> (+0.78%) ⬆️
appsec-integration 49.51% <ø> (+0.34%) ⬆️
appsec-kafka_ldapjs_lodash 67.72% <ø> (+0.93%) ⬆️
appsec-mongodb-core_mongoose_mysql 70.92% <ø> (+0.81%) ⬆️
appsec-next 57.23% <ø> (+0.75%) ⬆️
appsec-node-serialize_passport_postgres 70.38% <ø> (+0.83%) ⬆️
appsec-sourcing_stripe_template 68.79% <ø> (+0.84%) ⬆️
debugger 69.38% <ø> (+0.41%) ⬆️
instrumentations-bucket-0 56.21% <ø> (+1.30%) ⬆️
instrumentations-bucket-1 64.43% <ø> (+1.07%) ⬆️
instrumentations-bucket-10 65.57% <ø> (+0.96%) ⬆️
instrumentations-bucket-11 66.21% <ø> (+0.94%) ⬆️
instrumentations-bucket-12 55.99% <ø> (+1.20%) ⬆️
instrumentations-bucket-13 56.17% <ø> (+1.08%) ⬆️
instrumentations-bucket-14 56.01% <ø> (+1.10%) ⬆️
instrumentations-bucket-2 57.06% <ø> (+1.06%) ⬆️
instrumentations-bucket-3 57.84% <ø> (+1.02%) ⬆️
instrumentations-bucket-4 63.88% <ø> (+1.07%) ⬆️
instrumentations-bucket-5 50.14% <ø> (+0.93%) ⬆️
instrumentations-bucket-6 65.76% <ø> (+1.09%) ⬆️
instrumentations-bucket-7 56.22% <ø> (+1.17%) ⬆️
instrumentations-bucket-8 63.21% <ø> (+1.01%) ⬆️
instrumentations-bucket-9 58.63% <ø> (+0.98%) ⬆️
instrumentations-instrumentation-couchbase 54.80% <ø> (+1.30%) ⬆️
instrumentations-integration-esbuild 34.42% <ø> (+0.35%) ⬆️
llmobs-ai_anthropic_bedrock 66.92% <ø> (+0.78%) ⬆️
llmobs-bucket-1 65.02% <ø> (+0.75%) ⬆️
llmobs-openai 66.96% <ø> (+0.87%) ⬆️
llmobs-openai-agents_vertex-ai 64.14% <ø> (+0.83%) ⬆️
llmobs-sdk 76.49% <ø> (+1.16%) ⬆️
master-coverage 98.59% <ø> (?)
openfeature 59.77% <ø> (+0.25%) ⬆️
openfeature-unit 58.44% <ø> (+1.14%) ⬆️
platform-core_esbuild_instrumentations-misc 41.00% <ø> (+0.49%) ⬆️
platform-integration 64.99% <ø> (+0.78%) ⬆️
platform-shimmer_unit-guardrails_webpack 38.81% <ø> (+0.56%) ⬆️
plugins-browser-bunyan_bullmq_cassandra 66.52% <ø> (+1.06%) ⬆️
plugins-bucket-0 61.11% <ø> (+0.94%) ⬆️
plugins-bucket-1 58.31% <ø> (+0.89%) ⬆️
plugins-bucket-11 66.99% <ø> (+1.01%) ⬆️
plugins-bucket-18 66.42% <ø> (+0.91%) ⬆️
plugins-bucket-19 65.21% <ø> (+1.12%) ⬆️
plugins-bucket-20 67.03% <ø> (+1.07%) ⬆️
plugins-bucket-4 61.30% <ø> (+1.00%) ⬆️
plugins-cookie_cookie-parser_crypto 55.80% <ø> (+1.34%) ⬆️
plugins-fastify_fetch_fs 65.63% <ø> (+0.96%) ⬆️
plugins-generic-pool_google-cloud-pubsub_grpc 69.25% <ø> (+1.03%) ⬆️
plugins-handlebars_hapi_hono 63.84% <ø> (+1.14%) ⬆️
plugins-ioredis_knex_langgraph 61.82% <ø> (+1.03%) ⬆️
plugins-ldapjs_light-my-request_limitd-client 63.56% <ø> (+1.13%) ⬆️
plugins-lodash_mariadb_memcached 64.04% <ø> (+0.98%) ⬆️
plugins-moleculer_mongodb_mongodb-core 66.33% <ø> (+0.99%) ⬆️
plugins-mongoose_multer_mysql 64.14% <ø> (+0.98%) ⬆️
plugins-mysql2_nats_node-serialize 66.41% <ø> (+1.10%) ⬆️
plugins-opensearch_passport-http_pino 64.26% <ø> (+1.10%) ⬆️
plugins-postgres_process_pug 63.29% <ø> (+1.03%) ⬆️
plugins-redis_router_sequelize 66.91% <ø> (+1.00%) ⬆️
plugins-test-and-upstream-rhea_undici_url 66.24% <ø> (+0.96%) ⬆️
plugins-valkey_vm_winston 62.60% <ø> (+1.02%) ⬆️
plugins-ws 64.70% <ø> (+1.15%) ⬆️
profiling 66.41% <ø> (+0.83%) ⬆️
serverless-aws-sdk-aws-sdk 55.50% <ø> (+0.83%) ⬆️
serverless-aws-sdk-base-inject-field 55.36% <ø> (+1.32%) ⬆️
serverless-aws-sdk-bedrockruntime 58.40% <ø> (+1.07%) ⬆️
serverless-aws-sdk-client 60.22% <ø> (+1.07%) ⬆️
serverless-aws-sdk-dynamodb 59.28% <ø> (+1.00%) ⬆️
serverless-aws-sdk-eventbridge 57.74% <ø> (+0.83%) ⬆️
serverless-aws-sdk-kinesis 63.32% <ø> (+0.98%) ⬆️
serverless-aws-sdk-lambda 61.25% <ø> (+0.99%) ⬆️
serverless-aws-sdk-s3 59.33% <ø> (+0.98%) ⬆️
serverless-aws-sdk-serverless-peer-service 63.90% <ø> (+0.99%) ⬆️
serverless-aws-sdk-sns 64.22% <ø> (+0.99%) ⬆️
serverless-aws-sdk-sqs 64.68% <ø> (+0.99%) ⬆️
serverless-aws-sdk-stepfunctions 59.19% <ø> (+1.03%) ⬆️
serverless-aws-sdk-util 56.00% <ø> (+1.30%) ⬆️
serverless-bucket-0 57.94% <ø> (+1.04%) ⬆️
serverless-bucket-1 63.92% <ø> (+1.07%) ⬆️
test-optimization-cucumber 70.77% <ø> (+0.54%) ⬆️
test-optimization-cypress 64.93% <ø> (+0.49%) ⬆️
test-optimization-jest 72.30% <ø> (+0.45%) ⬆️
test-optimization-mocha 72.22% <ø> (+0.56%) ⬆️
test-optimization-playwright-playwright-atr 59.92% <ø> (+0.54%) ⬆️
test-optimization-playwright-playwright-efd 60.56% <ø> (+0.52%) ⬆️
test-optimization-playwright-playwright-final-status 60.17% <ø> (+0.52%) ⬆️
test-optimization-playwright-playwright-impacted-tests 60.28% <ø> (+0.66%) ⬆️
test-optimization-playwright-playwright-reporting 61.26% <ø> (+0.43%) ⬆️
test-optimization-playwright-playwright-test-management 61.35% <ø> (+0.43%) ⬆️
test-optimization-playwright-playwright-test-span 59.96% <ø> (+0.48%) ⬆️
test-optimization-selenium 58.97% <ø> (+0.35%) ⬆️
test-optimization-testopt 61.91% <ø> (+0.39%) ⬆️
test-optimization-vitest 72.92% <ø> (+0.52%) ⬆️
test-optimization-vitest-browser 58.80% <ø> (+0.42%) ⬆️
test-optimization-webdriverio 65.66% <ø> (+0.74%) ⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

@pabloerhard
pabloerhard marked this pull request as ready for review August 26, 2026 17:44
@pabloerhard
pabloerhard requested review from a team as code owners August 26, 2026 17:44
@pabloerhard
pabloerhard requested a review from bojbrook August 26, 2026 17:44

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 86f1a28d3c

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread .github/CODEOWNERS Outdated
@pabloerhard
pabloerhard force-pushed the pabloerhard/refactor-agents-md branch from 86f1a28 to 7cdfd8b Compare August 26, 2026 18:32
@wconti27

Copy link
Copy Markdown
Contributor

Changes look good to me from IDM POV!

@wconti27

Copy link
Copy Markdown
Contributor

I'm very curous here if we have any measurements around agent performance related to .md agent context changes. Bc otherwise these changes are very based on vibes

@rochdev rochdev left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Left a few comments but nothing major, could be addressed later or as a follow-up post-approval.

Comment thread .agents/skills/apm-integrations/references/new-integration-guide.md Outdated
Comment thread .agents/skills/apm-integrations/references/new-integration-guide.md Outdated

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 43bb04f3fe

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread AGENTS.md Outdated

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: e7d4a0f3e4

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread .agents/skills/apm-integrations/references/new-integration-guide.md
Comment thread .agents/skills/apm-integrations/references/testing.md Outdated

@bengl bengl left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

This needs to come with some sort of check that AGENTS.md is sufficiently streamlined.

@pabloerhard
pabloerhard force-pushed the pabloerhard/refactor-agents-md branch from 7182a5f to 60d1b73 Compare August 27, 2026 17:49
@pabloerhard
pabloerhard requested a review from bengl August 27, 2026 17:53
@bengl
bengl enabled auto-merge (squash) August 27, 2026 17:56
@pabloerhard
pabloerhard force-pushed the pabloerhard/refactor-agents-md branch from 60d1b73 to 04fd80f Compare August 27, 2026 18:57
@pabloerhard
pabloerhard force-pushed the pabloerhard/refactor-agents-md branch from 04fd80f to ba99631 Compare August 27, 2026 19:30

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: ba99631f01

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread .agents/skills/flaky-test-investigation/SKILL.md Outdated
@pabloerhard
pabloerhard requested a review from a team August 27, 2026 21:02
@pabloerhard
pabloerhard requested review from a team and mabdinur and removed request for a team August 27, 2026 21:13
@bengl
bengl merged commit 4d9ea3d into master Aug 28, 2026
686 checks passed
@bengl
bengl deleted the pabloerhard/refactor-agents-md branch August 28, 2026 21:49
This was referenced Aug 31, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants