fix(llmobs): format AI SDK rich tool results by BridgeAR · Pull Request #9399 · DataDog/dd-trace-js

BridgeAR · 2026-07-16T02:49:36Z

Summary

AI SDK multipart and error tool outputs were treated as unparsable. This preserves text, emits metadata-free summaries for rich file and image parts, and keeps malformed or circular telemetry inputs from throwing.

Fixes: #9395

dd-octo-sts · 2026-07-16T02:50:26Z

Overall package size

Self size: 7.48 MB
Deduped: 8.14 MB
No deduping: 8.14 MB

Dependency sizes

| name | version | self size | total size | |------|---------|-----------|------------| | import-in-the-middle | 3.3.2 | 124.41 kB | 440.65 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}

pr-commenter · 2026-07-16T03:00:22Z

Benchmarks

Benchmark execution time: 2026-07-21 01:30:52

Comparing candidate commit d91d6d8 in PR branch BridgeAR/2026-07-16-fix-ai-tool-results with baseline commit c8e8699 in branch master.

📊 Benchmarking dashboard

Found 0 performance improvements and 0 performance regressions! Performance is the same for 2321 metrics, 37 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 execution_time [-211.888ms; +204.004ms] or [-7.970%; +7.674%]

scenario:appsec-appsec-enabled-26

unstable execution_time [-238.715ms; +236.240ms] or [-9.351%; +9.254%]

scenario:appsec-appsec-enabled-with-attacks-24

unstable execution_time [-168.824ms; +154.477ms] or [-5.460%; +4.996%]

scenario:appsec-appsec-enabled-with-attacks-26

unstable execution_time [-191.805ms; +204.872ms] or [-6.508%; +6.951%]

scenario:appsec-control-20

unstable execution_time [-124720.953µs; +124948.753µs] or [-7.576%; +7.590%]

scenario:appsec-control-24

unstable execution_time [-114526.148µs; +114674.814µs] or [-9.216%; +9.228%]

scenario:appsec-control-26

unstable execution_time [-123.119ms; +120.580ms] or [-9.953%; +9.748%]

scenario:appsec-iast-no-vulnerability-iast-enabled-always-active-20

unstable execution_time [-18.107ms; +14.299ms] or [-6.970%; +5.504%]

scenario:appsec-iast-no-vulnerability-iast-enabled-default-config-20

unstable execution_time [-15.769ms; +12.662ms] or [-6.171%; +4.955%]

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

unstable cpu_user_time [-1986.122ms; +3192.797ms] or [-23.912%; +38.440%]
unstable execution_time [-2002.386ms; +3206.113ms] or [-22.173%; +35.502%]
unstable instructions [-17.1G instructions; +27.2G instructions] or [-25.201%; +40.164%]
unstable max_rss_usage [-8.165MB; +13.326MB] or [-5.201%; +8.489%]
unstable throughput [-848.393op/s; +527.386op/s] or [-23.261%; +14.460%]

scenario:debugger-line-probe-with-snapshot-default-26

unstable cpu_user_time [-2.936s; +5.691s] or [-28.063%; +54.402%]
unstable execution_time [-2.887s; +5.721s] or [-25.813%; +51.148%]
unstable instructions [-26.2G instructions; +50.7G instructions] or [-29.751%; +57.639%]
unstable max_rss_usage [-9.539MB; +18.230MB] or [-5.869%; +11.215%]
unstable throughput [-1131.165op/s; +513.821op/s] or [-36.789%; +16.711%]

scenario:debugger-line-probe-with-snapshot-minimal-26

unstable cpu_user_time [-2697.926ms; +4214.287ms] or [-28.305%; +44.213%]
unstable execution_time [-2831.902ms; +4346.768ms] or [-27.620%; +42.395%]
unstable instructions [-23.1G instructions; +36.9G instructions] or [-29.057%; +46.391%]
unstable max_rss_usage [-8.304MB; +13.237MB] or [-5.202%; +8.292%]
unstable throughput [-835.134op/s; +560.646op/s] or [-25.762%; +17.295%]

scenario:dogstatsd-with-tags-20

unstable cpu_user_time [-393.085ms; +350.559ms] or [-8.242%; +7.351%]
unstable execution_time [-391.279ms; +355.536ms] or [-8.072%; +7.335%]
unstable throughput [-110137.328op/s; +140622.990op/s] or [-6.341%; +8.096%]

scenario:plugin-claude-agent-sdk-delayed-noisy-hook-indexed-26

unstable execution_time [-10.688ms; +19.377ms] or [-3.798%; +6.885%]

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

unstable max_rss_usage [-19.360MB; +32.135MB] or [-5.168%; +8.578%]

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

unstable max_rss_usage [-5.384MB; +11.415MB] or [-4.191%; +8.885%]

scenario:plugin-graphql-long-with-depth-off-26

unstable max_rss_usage [-25.532MB; +17.377MB] or [-11.518%; +7.839%]

scenario:plugin-graphql-long-with-depth-on-max-20

unstable cpu_user_time [-562.525ms; +589.353ms] or [-4.892%; +5.125%]
unstable execution_time [-577.184ms; +609.557ms] or [-4.915%; +5.191%]
unstable throughput [-3.571op/s; +3.380op/s] or [-5.211%; +4.932%]

scenario:plugin-grpc-mixed-26

unstable cpu_user_time [-159.364ms; +268.177ms] or [-4.817%; +8.105%]
unstable execution_time [-163.591ms; +266.235ms] or [-4.813%; +7.833%]

scenario:test-optimization-large-suite-20

unstable max_rss_usage [-5363.951KB; +5162.951KB] or [-6.638%; +6.390%]

datadog-prod-us1-5 · 2026-07-16T03:03:01Z

Tests

⚠️ Warnings

🚦 2 Pipeline jobs failed

All Green | all-green

Platform | integration-guardrails (22.0.0)

ℹ️ Info

No other issues found (see more)

🧪 All tests passed
❄️ No new flaky tests detected

🔄 Datadog retried 1 test - 1 passed on retry

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 98.42% (+0.00%)

Useful? React with 👍 / 👎

_{This comment will be updated automatically if new data arrives.

🔗 Commit SHA: d91d6d8 | Docs | Datadog PR Page | Give us feedback!}

codecov · 2026-07-16T05:04:10Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.42%. Comparing base (c8e8699) to head (d91d6d8).

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #9399   +/-   ##
=======================================
  Coverage   98.42%   98.42%           
=======================================
  Files         938      938           
  Lines      126209   126290   +81     
  Branches    10735    10666   -69     
=======================================
+ Hits       124219   124300   +81     
  Misses       1990     1990

Flag	Coverage Δ
aiguard	`57.80% <ø> (-0.03%)`	⬇️
aiguard-integration	`56.20% <36.53%> (-0.03%)`	⬇️
apm-bucket-0	`57.65% <ø> (-0.03%)`	⬇️
apm-bucket-1	`63.72% <ø> (-0.03%)`	⬇️
apm-bucket-2	`62.74% <ø> (-0.03%)`	⬇️
apm-bucket-3	`60.10% <ø> (-0.03%)`	⬇️
apm-capabilities-tracing	`64.18% <36.53%> (-0.03%)`	⬇️
apm-integrations-aerospike	`56.65% <ø> (-0.03%)`	⬇️
apm-integrations-confluentinc-kafka-javascript	`61.62% <ø> (-0.03%)`	⬇️
apm-integrations-couchbase	`57.08% <ø> (-0.04%)`	⬇️
apm-integrations-http	`62.79% <ø> (-0.03%)`	⬇️
apm-integrations-kafkajs	`62.17% <ø> (-0.03%)`	⬇️
apm-integrations-next	`59.20% <ø> (-0.03%)`	⬇️
apm-integrations-prisma	`58.68% <ø> (-0.03%)`	⬇️
appsec	`72.94% <ø> (-0.07%)`	⬇️
appsec-express_fastify_graphql	`70.45% <ø> (-0.02%)`	⬇️
appsec-integration	`51.44% <ø> (+<0.01%)`	⬆️
appsec-kafka_ldapjs_lodash	`63.99% <ø> (-0.03%)`	⬇️
appsec-mongodb-core_mongoose_mysql	`67.73% <ø> (-0.02%)`	⬇️
appsec-next	`57.51% <ø> (-0.02%)`	⬇️
appsec-node-serialize_passport_postgres	`67.40% <ø> (-0.02%)`	⬇️
appsec-sourcing_stripe_template	`65.73% <ø> (-0.02%)`	⬇️
debugger	`64.96% <ø> (-0.03%)`	⬇️
instrumentations-bucket-0	`51.61% <ø> (-0.03%)`	⬇️
instrumentations-bucket-1	`60.21% <ø> (-0.03%)`	⬇️
instrumentations-bucket-10	`61.79% <ø> (-0.03%)`	⬇️
instrumentations-bucket-11	`56.67% <ø> (-0.03%)`	⬇️
instrumentations-bucket-12	`51.94% <ø> (-0.03%)`	⬇️
instrumentations-bucket-13	`51.99% <ø> (-0.03%)`	⬇️
instrumentations-bucket-2	`53.28% <ø> (-0.03%)`	⬇️
instrumentations-bucket-3	`53.58% <ø> (-0.03%)`	⬇️
instrumentations-bucket-4	`59.21% <ø> (-0.03%)`	⬇️
instrumentations-bucket-5	`49.98% <ø> (-0.03%)`	⬇️
instrumentations-bucket-6	`60.93% <ø> (-0.03%)`	⬇️
instrumentations-bucket-7	`58.36% <ø> (-0.03%)`	⬇️
instrumentations-bucket-8	`59.44% <ø> (-0.03%)`	⬇️
instrumentations-bucket-9	`51.99% <ø> (-0.03%)`	⬇️
instrumentations-instrumentation-couchbase	`51.04% <ø> (-0.03%)`	⬇️
instrumentations-instrumentation-zlib	`51.61% <ø> (-0.03%)`	⬇️
instrumentations-integration-esbuild	`34.12% <ø> (ø)`
llmobs-ai_anthropic_bedrock	`62.64% <100.00%> (+0.04%)`	⬆️
llmobs-bucket-1	`61.84% <ø> (-0.02%)`	⬇️
llmobs-openai	`62.71% <ø> (-0.03%)`	⬇️
llmobs-sdk	`66.23% <ø> (-0.03%)`	⬇️
llmobs-vertex-ai	`59.18% <ø> (-0.03%)`	⬇️
master-coverage	`98.42% <100.00%> (?)`
openfeature	`53.98% <ø> (+<0.01%)`	⬆️
openfeature-unit	`52.72% <ø> (-0.03%)`	⬇️
platform-core_esbuild_instrumentations-misc	`40.37% <ø> (-0.02%)`	⬇️
platform-integration	`61.51% <ø> (ø)`
platform-shimmer_unit-guardrails_webpack	`39.06% <ø> (-0.02%)`	⬇️
plugins-bucket-0	`57.06% <ø> (-0.03%)`	⬇️
plugins-bucket-1	`54.43% <ø> (ø)`
plugins-bucket-11	`61.85% <ø> (-0.48%)`	⬇️
plugins-bucket-17	`61.90% <ø> (?)`
plugins-bucket-18	`62.59% <ø> (+0.48%)`	⬆️
plugins-bucket-19	`61.69% <ø> (+1.56%)`	⬆️
plugins-bucket-20	`64.33% <ø> (+2.24%)`	⬆️
plugins-bucket-4	`58.59% <ø> (-0.03%)`	⬇️
plugins-bullmq_cassandra_cookie	`61.80% <ø> (-0.03%)`	⬇️
plugins-cookie-parser_crypto_dd-trace-api	`56.75% <ø> (-0.03%)`	⬇️
plugins-fetch_fs_generic-pool	`58.86% <ø> (-0.03%)`	⬇️
plugins-google-cloud-pubsub_grpc_handlebars	`64.76% <ø> (-0.03%)`	⬇️
plugins-hapi_hono_ioredis	`60.27% <ø> (-0.03%)`	⬇️
plugins-jest_knex_langgraph	`?`
plugins-jest_langgraph_ldapjs	`55.57% <ø> (?)`
plugins-ldapjs_light-my-request_limitd-client	`?`
plugins-light-my-request_limitd-client_lodash	`58.65% <ø> (?)`
plugins-lodash_mariadb_memcached	`?`
plugins-mariadb_memcached_mercurius	`61.61% <ø> (?)`
plugins-moleculer_mongodb_mongodb-core	`?`
plugins-mongodb_mongodb-core_mongoose	`59.71% <ø> (?)`
plugins-mongoose_multer_mysql	`?`
plugins-multer_mysql_mysql2	`58.44% <ø> (?)`
plugins-mysql2_nats_node-serialize	`?`
plugins-nats_node-serialize_opensearch	`60.83% <ø> (?)`
plugins-opensearch_passport-http_pino	`?`
plugins-passport-http_pino_postgres	`58.66% <ø> (?)`
plugins-postgres_process_pug	`?`
plugins-process_pug_redis	`57.70% <ø> (?)`
plugins-redis_router_sequelize	`?`
plugins-test-and-upstream-rhea_undici_url	`?`
plugins-undici_url_valkey	`58.59% <ø> (?)`
plugins-valkey_vm_winston	`?`
plugins-vm_winston_ws	`59.96% <ø> (?)`
plugins-ws	`?`
profiling	`62.22% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-aws-sdk	`55.08% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-bedrockruntime	`54.77% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-client	`56.44% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-dynamodb	`55.68% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-eventbridge	`49.38% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-kinesis	`59.38% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-lambda	`57.42% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-s3	`55.62% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-serverless-peer-service	`59.79% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-sns	`60.18% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-sqs	`60.61% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-stepfunctions	`55.60% <ø> (-0.03%)`	⬇️
serverless-aws-sdk-util	`51.49% <ø> (-0.03%)`	⬇️
serverless-bucket-0	`54.32% <ø> (ø)`
serverless-bucket-1	`59.29% <ø> (-0.03%)`	⬇️
test-optimization-cucumber	`71.82% <ø> (+0.04%)`	⬆️
test-optimization-cypress	`65.80% <ø> (+0.07%)`	⬆️
test-optimization-jest	`73.30% <ø> (+<0.01%)`	⬆️
test-optimization-mocha	`73.47% <ø> (+0.04%)`	⬆️
test-optimization-playwright-playwright-atr	`60.42% <ø> (-0.02%)`	⬇️
test-optimization-playwright-playwright-efd	`60.60% <ø> (-0.01%)`	⬇️
test-optimization-playwright-playwright-final-status	`60.57% <ø> (-0.02%)`	⬇️
test-optimization-playwright-playwright-impacted-tests	`60.30% <ø> (+0.15%)`	⬆️
test-optimization-playwright-playwright-reporting	`60.20% <ø> (-0.02%)`	⬇️
test-optimization-playwright-playwright-test-management	`61.20% <ø> (+0.18%)`	⬆️
test-optimization-playwright-playwright-test-span	`60.33% <ø> (-0.06%)`	⬇️
test-optimization-selenium	`59.82% <ø> (-0.17%)`	⬇️
test-optimization-testopt	`58.46% <ø> (+0.08%)`	⬆️
test-optimization-vitest	`70.22% <ø> (+0.01%)`	⬆️

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.

sabrenner · 2026-07-20T14:22:23Z

+}
+
+/**
+ * @param {unknown} value


nit - we could probably better define this type at least with the knowledge we have below on the properties we access from each part

sabrenner · 2026-07-20T14:25:28Z

i would prefer we just write new tests in the existing test file for a more "end-to-end" testing experience. especially since the underlying utilities are also relying on object shapes from the actual ai sdk

AI SDK tool outputs can contain multipart content and error variants. Preserve text, replace file and image payloads with safe summaries, and fall back without throwing on malformed values. Fixes: #9395

AI SDK 7 emits generic file and provider-reference parts that were reported as unparsable. Model the supported output shapes locally and exercise content outputs through both telemetry paths so SDK shape changes fail on the user path.

Adding toModelOutput changed the second provider request without a matching cassette, so every AI SDK v5-v7 matrix forwarded to OpenAI with the fake key. Keep the recorded workflow payload stable while formatter coverage pins the rich result shapes and schema precedence.

github-actions Bot added the semver-patch label Jul 16, 2026

sabrenner reviewed Jul 20, 2026

View reviewed changes

BridgeAR force-pushed the BridgeAR/2026-07-16-fix-ai-tool-results branch from aaa217b to aab8927 Compare July 20, 2026 18:14

BridgeAR added 3 commits July 21, 2026 03:17

fix(llmobs): format AI SDK rich tool results

988d130

AI SDK tool outputs can contain multipart content and error variants. Preserve text, replace file and image payloads with safe summaries, and fall back without throwing on malformed values. Fixes: #9395

fix(llmobs): support AI SDK file result parts

34f4ea3

AI SDK 7 emits generic file and provider-reference parts that were reported as unparsable. Model the supported output shapes locally and exercise content outputs through both telemetry paths so SDK shape changes fail on the user path.

BridgeAR force-pushed the BridgeAR/2026-07-16-fix-ai-tool-results branch from aab8927 to d91d6d8 Compare July 21, 2026 01:19

Uh oh!

Conversation

BridgeAR commented Jul 16, 2026

Summary

Uh oh!

dd-octo-sts Bot commented Jul 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Overall package size

Uh oh!

pr-commenter Bot commented Jul 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Benchmarks

Explanation

More details about the CI and significant changes

Unstable benchmarks

scenario:appsec-appsec-enabled-24

scenario:appsec-appsec-enabled-26

scenario:appsec-appsec-enabled-with-attacks-24

scenario:appsec-appsec-enabled-with-attacks-26

scenario:appsec-control-20

scenario:appsec-control-24

scenario:appsec-control-26

scenario:appsec-iast-no-vulnerability-iast-enabled-always-active-20

scenario:appsec-iast-no-vulnerability-iast-enabled-default-config-20

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

scenario:debugger-line-probe-with-snapshot-default-26

scenario:debugger-line-probe-with-snapshot-minimal-26

scenario:dogstatsd-with-tags-20

scenario:plugin-claude-agent-sdk-delayed-noisy-hook-indexed-26

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

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

scenario:plugin-graphql-long-with-depth-off-26

scenario:plugin-graphql-long-with-depth-on-max-20

scenario:plugin-grpc-mixed-26

scenario:test-optimization-large-suite-20

Uh oh!

datadog-prod-us1-5 Bot commented Jul 16, 2026 • edited by datadog-official Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

⚠️ Warnings

ℹ️ Info

Uh oh!

codecov Bot commented Jul 16, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

sabrenner Jul 20, 2026

Choose a reason for hiding this comment

Uh oh!

sabrenner Jul 20, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

dd-octo-sts Bot commented Jul 16, 2026 •

edited

Loading

pr-commenter Bot commented Jul 16, 2026 •

edited

Loading

datadog-prod-us1-5 Bot commented Jul 16, 2026 •

edited by datadog-official Bot

Loading

codecov Bot commented Jul 16, 2026 •

edited

Loading