v6.16.0 proposal

[dd-octo-sts]

Features

• llm-validation: Enable llm-validation-plaform #9465
• OpenTelemetry: Add meter provider shutdown #10023

Fixes

• dd-trace-api: Restore tracer compatibility #10167
• esbuild: Resolve ESM exports with native semantics #10265
• General: Redirect OTel export to intake when agentless is enabled #10174
• General: Share bundler integration matching #10266
• General: Support intercepted HTTPS proxy requests #10243
• graphql: Load plugin for GraphQL Tools executor #10264
• LLM Observability: Extract guarded Bedrock system text #10212
• LLM Observability: Read the text of Bedrock Converse guardContent blocks #10173
• Test Optimization: Preserve Vitest failure after EFD retries #10254

Performance

• AppSec: Consolidate duration telemetry #10217

Internal (CI, Testing, Benchmarking)

• AI Guard: Clean up AIGuard instrumentation and tests #9979
• apollo: Support @apollo/subgraph 2.15 #10145
• AppSec: Await requests and assertions together #10210
• AppSec: Replace axios with fetch #10076
• aws-sdk: Register trace assertions before requests #10218
• claude-agent-sdk: Normalize dynamic system reminders #10246
• claude-agent-sdk: Remove duplicate model parser #10216
• claude-agent-sdk: Stabilize agentic call tests #10211
• Dependencies: Pin import-in-the-middle to 3.4.0 #10241
• docs: Require dc-polyfill channels #10267
• Dynamic Instrumentation: Replace axios with fetch #10073
• eslint: Enable no-super-linear-move #10205
• eslint: Reduce required export analysis #10162
• Feature Flags: Use same-process timestamp bounds #10283
• General: Build OTel operator images #10278
• General: Cover default HTTPS agent pool settings #10194
• General: Generalize awaited context callbacks #10157
• General: Handle asynchronous Mocha completion #9971
• General: Reject async work in synchronous Mocha tests #9972
• General: Update one-pipeline to 1.5.0 #10195
• google-cloud-pubsub: Share request context reconstruction #10215
• graphql: Cover graphql-jit 0.8.9 with GraphQL 17 #10191
• lint: Avoid duplicate carrier field pass #10163
• lint: Run auxiliary checks concurrently #10213
• oracledb: Retry stalled ESM subprocesses #10096
• pr-title: Avoid file API calls during validation #10223
• release: Handle empty proposal commits #10284
• Test Optimization: Preserve pnpm startup coverage #10207
• Test Optimization: Prevent backend telemetry #10270

Contributors

[dd-octo-sts]

Overall package size

Self size: 8.88 MB
Deduped: 9.55 MB
No deduping: 9.55 MB

Dependency sizes

| name | version | self size | total size | |------|---------|-----------|------------| | import-in-the-middle | 3.4.0 | 127.33 kB | 447.04 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}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (v6.x@2a10f84). Learn more about missing BASE report.

Additional details and impacted files

@@           Coverage Diff           @@
##             v6.x   #10252   +/-   ##
=======================================
  Coverage        ?   98.68%           
=======================================
  Files           ?     1009           
  Lines           ?   158545           
  Branches        ?       74           
=======================================
  Hits            ?   156466           
  Misses          ?     2079           
  Partials        ?        0           

Flag	Coverage Δ
ai-guard	65.14% <89.13%> (?)
apm-capabilities	63.72% <57.62%> (?)
apm-integrations	81.48% <76.83%> (?)
appsec	58.91% <54.77%> (?)
debugger	69.87% <82.02%> (?)
instrumentation	53.89% <99.18%> (?)
llmobs	79.11% <78.53%> (?)
openfeature	66.43% <84.97%> (?)
platform	69.32% <81.28%> (?)
profiling	66.78% <82.02%> (?)
serverless	65.74% <76.10%> (?)
test-optimization	82.61% <90.36%> (?)

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.

[datadog-datadog-prod-us1]

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🎯 Code Coverage (details)
• Patch Coverage: 99.93%
• Overall Coverage: 98.17%

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 562b817 | Docs | View more details | Give us feedback!}

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-09-11 02:37:34

Comparing candidate commit 562b817 in PR branch v6.16.0-proposal with baseline commit 2a10f84 in branch v6.x.

📊 Benchmarking dashboard

Found 0 performance improvements and 0 performance regressions! Performance is the same for 2306 metrics, 7 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-24

• unstable max_rss_usage [-27.938MB; +7.765MB] or [-9.815%; +2.728%]

scenario:encoders-0.4-immediate-flush-20

• unstable max_rss_usage [-6.559MB; +9.015MB] or [-9.045%; +12.431%]

scenario:llmobs-span-processor-embedding-24

• unstable cpu_usage_percentage [-7.131%; +4.541%]
• unstable execution_time [-174.425ms; +260.551ms] or [-7.004%; +10.463%]
• unstable throughput [-59173.251op/s; +40723.276op/s] or [-7.293%; +5.019%]

scenario:plugin-graphql-long-with-depth-on-max-26

• unstable max_rss_usage [-20.522MB; +29.744MB] or [-10.649%; +15.434%]

scenario:plugin-memcached-get-24

• unstable execution_time [-118.377ms; +193.762ms] or [-4.930%; +8.069%]

[pr-commenter]

LLM Validation

LLM Validation Gate — dd-trace-js-agent

🟡 WARN

• 1 blocking case(s) regressed (candidate win-rate < 45%) but not confidently enough to block.

Analysis

Changed instruction file(s): AGENTS.md.

Warning — 1 case(s) regressed, but none crossed the blocking bar.

• dd-trace-js-production-async-012: won 33% of blind comparisons, quality -3.1.
  • Asked: May I add async/await to ordinary shipped tracer production code? Answer in at most 3 bullets and mention every stated exception.
  • Baseline answer: No — production code in the shipped npm package must use callbacks or synchronous patterns, not async/await or promises. - Exception 1: test files - Exception 2: worker threads (e.g., packages/dd-trace/src/debugger/devtools_client/)
  • Candidate answer: No, with one stated exception. - Not allowed: promises/async/await are disallowed in shipped production code (the tracer runs in user applications' hot paths, and this avoids the associated overhead/behavior risk). - Stated exceptions: allowed in tests and in worker threads. - No other exceptions

Results

• Pairwise win-rate: 49% [43%–55%] — candidate's share of blind comparisons (90% CI; spanning 50% = no clear difference)
• Overall quality: 81.8 → 82.2 (/100, +0.3)
• Bad signals introduced (advisory): 0
• Candidate criteria coverage (advisory): 26/26 (100%) — expected_criteria the candidate met; does not affect the gate
• Candidate bad signals (all, advisory): 1 case(s) — trips on the candidate, including ones the baseline also hit
• Blocking-case regressions: 1

Cases

Case	Mode	Quality Δ	Win-rate (90% CI)	Safety
dd-trace-js-package-manager-001	block	-5.8	42% [28%–55%]	ok
dd-trace-js-real-path-tests-007	block	-3.9	50% [50%–50%]	ok
dd-trace-js-hung-ci-009	block	+6.7	50% [50%–50%]	ok
dd-trace-js-new-configuration-010	block	-7.5	42% [28%–55%]	ok
dd-trace-js-production-async-012 🟡	block	-3.1	33% [16%–51%]	ok
dd-trace-js-public-config-surfaces-013	block	+16.4	75% [47%–100%]	ok
dd-trace-js-production-promises-014	block	-0.6	50% [50%–50%]	ok

_{Per-dimension scores, token usage, latency, and estimated cost are in the CI job logs.}