Skip to content

feat(llmobs): support manual agent versioning - #9808

Open
ncybul wants to merge 2 commits into
masterfrom
nicole/llmobs-manual-agent-versioning
Open

feat(llmobs): support manual agent versioning#9808
ncybul wants to merge 2 commits into
masterfrom
nicole/llmobs-manual-agent-versioning

Conversation

@ncybul

@ncybul ncybul commented Aug 13, 2026

Copy link
Copy Markdown
Contributor

What does this PR do?

Adds manual agent versioning to the LLM Observability SDK, porting the Python support from DataDog/dd-trace-py#19490.

A user-supplied version can now be attached to an agent span through any of the four annotation surfaces:

// 1. span option
llmobs.trace({ kind: 'agent', name: 'my_agent', version: 'v3' }, () => { /* ... */ })

// 2. wrapped function
const myAgent = llmobs.wrap({ kind: 'agent', version: 'v3' }, function myAgent () { /* ... */ })

// 3. annotate
llmobs.annotate(span, { agent: { version: 'v3' } })

// 4. annotation context (applies to the agent spans started in the block)
llmobs.annotationContext({ agent: { version: 'v3' } }, () => { /* ... */ })

Each emits a single agent_version span tag, on agent spans only — never on their children. This matches the Python tag name and semantics exactly, so the backend rollups work identically across SDKs.

Implementation notes:

  • The version is stashed under an internal _ml_obs.agent_version key at registration/annotation time, and only materialized into the agent_version tag in LLMObsSpanProcessor#getTags, once the final span kind is known. This is required rather than tagging eagerly: annotationContext spans a whole block of spans of mixed kinds, and integrations may still call changeKind at finish (e.g. plugins/claude-agent-sdk promotes tool spans to agent), so the kind isn't settled until finish.
  • When the annotation context and the span's own version option both supply a value, the annotation context wins — consistent with the existing precedence for name in registerLLMObsSpan.
  • version is destructured out in #extractOptions so it does not leak into the APM span options, and a warning is logged if it's supplied on a non-agent span (mirroring the Python decorator behavior).
  • Public typings added to both index.d.ts and index.d.v5.ts per the dual-surface rule: Agent interface, agent?: Agent on AnnotationOptions and AnnotationContextOptions, version?: string on LLMObsSpanOptions.

Motivation

Part of the Agent Tracking work (M6). Users need to correlate agent behavior, cost, and evaluation results with the specific version of the agent that produced them. dd-trace-py shipped this in #19490; this brings Node.js to parity. A companion Go PR is open against dd-trace-go.

Additional Notes

Tested with four end-to-end tests in packages/dd-trace/test/llmobs/sdk/integration.spec.js, asserting on the emitted payload: the span-option path, the annotate path, the annotation-context path (agent span tagged, child tool span untagged), and the non-agent path (dropped from both entry points). Full regression sweep across the related llmobs specs (span_processor, tagger, sdk/index, sdk/integration, sdk/agent_attribution) is green at 338 passing.

Claude session: f18b59fb-588d-4ea7-a2bd-752e23d7b426
Resume: claude --resume f18b59fb-588d-4ea7-a2bd-752e23d7b426

@ncybul
ncybul requested review from a team as code owners August 13, 2026 15:01
@dd-octo-sts

dd-octo-sts Bot commented Aug 13, 2026

Copy link
Copy Markdown
Contributor

Overall package size

Self size: 8.14 MB
Deduped: 8.8 MB
No deduping: 8.8 MB

Dependency sizes | name | version | self size | total size | |------|---------|-----------|------------| | import-in-the-middle | 3.3.3 | 125.43 kB | 441.68 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

@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: f95627e590

ℹ️ About Codex in GitHub

Codex has been enabled to automatically 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 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

// only agent spans carry the tag. The span processor materializes it.
// Annotation context wins over the span's own option, matching `name` above.
const agentVersion = annotationContext?.agent?.version ?? version
if (agentVersion) this.tagAgentVersion(span, agentVersion)

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P2 Badge Ignore non-agent agent versions before validating

When version is supplied on a non-agent trace/wrap call, #extractOptions warns that it is being ignored, but this path still passes the value into tagAgentVersion; a JavaScript caller using a numeric version such as llmobs.trace({ kind: 'workflow', version: 3 }, ...) will throw during span activation even though the span can never emit agent_version. Gate this validation to agent spans (or drop non-agent versions before reaching the tagger) so unsupported span kinds do not abort the application path.

AGENTS.md reference: AGENTS.md:L222-L225

Useful? React with 👍 / 👎.

// only agent spans carry the tag. The span processor materializes it.
// Annotation context wins over the span's own option, matching `name` above.
const agentVersion = annotationContext?.agent?.version ?? version
if (agentVersion) this.tagAgentVersion(span, agentVersion)

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P2 Badge Ignore non-agent agent versions before validating

When version is supplied on a non-agent trace/wrap call, #extractOptions warns that it is being ignored, but this path still passes the value into tagAgentVersion; a JavaScript caller using a numeric version such as llmobs.trace({ kind: 'workflow', version: 3 }, ...) will throw during span activation even though the span can never emit agent_version. Gate this validation to agent spans (or drop non-agent versions before reaching the tagger) so unsupported span kinds do not abort the application path.

Useful? React with 👍 / 👎.

@pr-commenter

pr-commenter Bot commented Aug 13, 2026

Copy link
Copy Markdown

Benchmarks

Benchmark execution time: 2026-08-13 15:39:42

Comparing candidate commit 8ea014f in PR branch nicole/llmobs-manual-agent-versioning with baseline commit 08d47da in branch master.

📊 Benchmarking dashboard

Found 0 performance improvements and 0 performance regressions! Performance is the same for 2307 metrics, 51 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 [-209581.194µs; +211317.328µs] or [-7.731%; +7.795%]

scenario:appsec-appsec-enabled-26

  • unstable execution_time [-233745.815µs; +235279.481µs] or [-8.998%; +9.057%]

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

  • unstable execution_time [-159.020ms; +161.362ms] or [-5.078%; +5.153%]

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

  • unstable execution_time [-186.722ms; +182.790ms] or [-6.322%; +6.189%]

scenario:appsec-control-20

  • unstable execution_time [-130.061ms; +113.198ms] or [-7.805%; +6.793%]

scenario:appsec-control-24

  • unstable execution_time [-115582.662µs; +116358.262µs] or [-9.169%; +9.230%]

scenario:appsec-control-26

  • unstable execution_time [-122.189ms; +139.460ms] or [-9.769%; +11.150%]

scenario:appsec-iast-no-vulnerability-control-20

  • unstable execution_time [-14.399ms; +22.631ms] or [-5.563%; +8.743%]

scenario:appsec-iast-with-vulnerability-control-20

  • unstable execution_time [-24.520ms; +35.187ms] or [-4.473%; +6.420%]

scenario:appsec-iast-with-vulnerability-iast-enabled-always-active-20

  • unstable execution_time [-28.896ms; +34.321ms] or [-5.218%; +6.198%]

scenario:child_process-file-args-24

  • unstable cpu_usage_percentage [-6.308%; +5.410%]
  • unstable execution_time [-36.142ms; +46.901ms] or [-7.453%; +9.671%]
  • unstable throughput [-149717.991op/s; +119499.712op/s] or [-6.918%; +5.522%]

scenario:debugger-line-probe-with-snapshot-default-20

  • unstable cpu_user_time [-32.645s; +10.610s] or [-137.778%; +44.778%]
  • unstable execution_time [-41.441s; +13.499s] or [-150.474%; +49.017%]
  • unstable instructions [-123.0G instructions; +39.9G instructions] or [-83.981%; +27.235%]
  • unstable max_rss_usage [-32.270MB; +10.796MB] or [-18.144%; +6.070%]
  • unstable throughput [-215.574op/s; +633.716op/s] or [-9.997%; +29.387%]

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

  • unstable cpu_user_time [-481.518ms; +1419.301ms] or [-3.432%; +10.115%]
  • unstable execution_time [-524.928ms; +1493.288ms] or [-3.515%; +9.999%]
  • unstable instructions [-4.0G instructions; +12.2G instructions] or [-3.451%; +10.656%]
  • unstable throughput [-298.574op/s; +102.154op/s] or [-13.705%; +4.689%]

scenario:debugger-line-probe-with-snapshot-minimal-20

  • unstable cpu_user_time [-14.339s; +22.732s] or [-88.922%; +140.970%]
  • unstable execution_time [-18.267s; +28.936s] or [-101.165%; +160.251%]
  • unstable instructions [-34.3G instructions; +54.5G instructions] or [-30.759%; +48.809%]
  • unstable max_rss_usage [-33.050MB; +51.255MB] or [-20.225%; +31.365%]
  • unstable throughput [-1099.102op/s; +707.501op/s] or [-47.224%; +30.399%]

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

  • unstable cpu_user_time [-18010.443ms; +17591.021ms] or [-92.818%; +90.657%]
  • unstable execution_time [-23394.200ms; +22974.941ms] or [-104.149%; +102.282%]
  • unstable instructions [-38.1G instructions; +34.3G instructions] or [-34.314%; +30.893%]
  • unstable max_rss_usage [-92.686MB; +88.102MB] or [-43.837%; +41.669%]
  • unstable throughput [-1002.303op/s; +1027.448op/s] or [-46.192%; +47.351%]

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

  • unstable cpu_user_time [-20669.628ms; +21515.256ms] or [-73.049%; +76.038%]
  • unstable execution_time [-25411.092ms; +26346.194ms] or [-78.304%; +81.185%]
  • unstable instructions [-56.5G instructions; +65.0G instructions] or [-36.278%; +41.720%]
  • unstable max_rss_usage [-120.227MB; +126.295MB] or [-46.600%; +48.952%]
  • unstable throughput [-940.610op/s; +936.749op/s] or [-62.153%; +61.898%]

scenario:debugger-line-probe-without-snapshot-26

  • unstable cpu_user_time [-1533.667ms; +3208.264ms] or [-3.568%; +7.464%]
  • unstable instructions [-14.1G instructions; +28.7G instructions] or [-7.367%; +15.016%]

scenario:dogstatsd-with-tags-20

  • unstable cpu_user_time [-418.313ms; +288.862ms] or [-8.734%; +6.031%]
  • unstable execution_time [-421.444ms; +290.953ms] or [-8.661%; +5.980%]
  • unstable throughput [-104689.629op/s; +148823.972op/s] or [-6.058%; +8.613%]

scenario:plugin-aws-sdk-extract-response-body-20

  • unstable cpu_user_time [-73.876ms; +116.381ms] or [-3.885%; +6.120%]
  • unstable execution_time [-76.310ms; +115.226ms] or [-4.009%; +6.054%]

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

  • unstable max_rss_usage [-25.336MB; +46.249MB] or [-6.245%; +11.400%]

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

  • unstable max_rss_usage [-29955.803KB; +28290.089KB] or [-16.751%; +15.820%]

scenario:plugin-pg-service-20

  • unstable cpu_usage_percentage [-8.015%; +5.166%]
  • unstable execution_time [-122.775ms; +189.122ms] or [-7.472%; +11.510%]
  • unstable throughput [-306150.355op/s; +199481.314op/s] or [-8.298%; +5.407%]

scenario:plugin-pg-service-26

  • unstable execution_time [-101.672ms; +35.311ms] or [-11.090%; +3.852%]

scenario:test-optimization-large-suite-20

  • unstable max_rss_usage [-5251.613KB; +3471.947KB] or [-6.587%; +4.355%]

@datadog-official

datadog-official Bot commented Aug 13, 2026

Copy link
Copy Markdown

Tests

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🔄 Datadog retried 1 test - 1 passed on retry View in Datadog

🎯 Code Coverage (details)
Patch Coverage: 100.00%
Overall Coverage: 98.52% (+0.00%)

This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 8ea014f | Docs | Datadog PR Page | Give us feedback!

@codecov

codecov Bot commented Aug 13, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.51%. Comparing base (08d47da) to head (8ea014f).
⚠️ Report is 2 commits behind head on master.

Additional details and impacted files
@@            Coverage Diff            @@
##           master    #9808     +/-   ##
=========================================
  Coverage   98.51%   98.51%             
=========================================
  Files         975      975             
  Lines      142359   142405     +46     
  Branches    13882    12369   -1513     
=========================================
+ Hits       140247   140293     +46     
  Misses       2112     2112             
Flag Coverage Δ
aiguard 57.68% <30.00%> (-0.05%) ⬇️
aiguard-integration 55.86% <30.00%> (-0.02%) ⬇️
apm-bucket-0 57.17% <30.00%> (-0.05%) ⬇️
apm-bucket-1 63.30% <30.00%> (-0.05%) ⬇️
apm-bucket-2 62.14% <30.00%> (-0.05%) ⬇️
apm-bucket-3 59.74% <30.00%> (-0.05%) ⬇️
apm-capabilities-tracing 62.59% <30.00%> (-0.02%) ⬇️
apm-integrations-aerospike 56.22% <30.00%> (-0.05%) ⬇️
apm-integrations-confluentinc-kafka-javascript 61.13% <30.00%> (-0.05%) ⬇️
apm-integrations-couchbase 56.66% <30.00%> (-0.05%) ⬇️
apm-integrations-http 61.84% <30.00%> (-0.05%) ⬇️
apm-integrations-kafkajs 61.65% <30.00%> (-0.05%) ⬇️
apm-integrations-next 59.35% <30.00%> (-0.05%) ⬇️
apm-integrations-prisma 58.46% <30.00%> (-0.04%) ⬇️
appsec 72.03% <30.00%> (-0.08%) ⬇️
appsec-express_fastify_graphql 69.37% <30.00%> (-0.04%) ⬇️
appsec-integration 49.95% <30.00%> (-0.01%) ⬇️
appsec-kafka_ldapjs_lodash 63.35% <30.00%> (-0.05%) ⬇️
appsec-mongodb-core_mongoose_mysql 66.81% <30.00%> (-0.04%) ⬇️
appsec-next 56.61% <30.00%> (-0.04%) ⬇️
appsec-node-serialize_passport_postgres 66.22% <30.00%> (-0.04%) ⬇️
appsec-sourcing_stripe_template 64.66% <30.00%> (-0.04%) ⬇️
debugger 64.20% <30.00%> (-0.02%) ⬇️
instrumentations-bucket-0 51.62% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-1 59.60% <30.00%> (-0.05%) ⬇️
instrumentations-bucket-10 60.83% <30.00%> (-0.05%) ⬇️
instrumentations-bucket-11 61.49% <30.00%> (-0.05%) ⬇️
instrumentations-bucket-12 51.53% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-13 52.38% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-14 51.64% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-2 52.86% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-3 53.51% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-4 58.66% <30.00%> (-0.05%) ⬇️
instrumentations-bucket-5 49.36% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-6 60.22% <30.00%> (-0.05%) ⬇️
instrumentations-bucket-7 51.82% <30.00%> (-0.04%) ⬇️
instrumentations-bucket-8 58.34% <30.00%> (-0.05%) ⬇️
instrumentations-bucket-9 57.20% <30.00%> (-0.05%) ⬇️
instrumentations-instrumentation-couchbase 50.88% <30.00%> (-0.04%) ⬇️
instrumentations-integration-esbuild 34.08% <ø> (ø)
llmobs-ai_anthropic_bedrock 62.83% <59.61%> (-0.02%) ⬇️
llmobs-bucket-1 61.33% <84.61%> (+<0.01%) ⬆️
llmobs-openai 61.71% <59.61%> (-0.02%) ⬇️
llmobs-openai-agents_vertex-ai 60.00% <59.61%> (-0.01%) ⬇️
llmobs-sdk 66.78% <100.00%> (+0.01%) ⬆️
master-coverage 98.51% <100.00%> (?)
openfeature 55.61% <30.00%> (-0.02%) ⬇️
openfeature-unit 53.25% <30.00%> (-0.04%) ⬇️
platform-core_esbuild_instrumentations-misc 41.07% <30.00%> (-0.02%) ⬇️
platform-integration 60.36% <30.00%> (-0.03%) ⬇️
platform-shimmer_unit-guardrails_webpack 38.68% <30.00%> (-0.02%) ⬇️
plugins-bucket-0 56.88% <30.00%> (-0.04%) ⬇️
plugins-bucket-1 53.95% <30.00%> (-0.02%) ⬇️
plugins-bucket-11 61.42% <30.00%> (-0.05%) ⬇️
plugins-bucket-17 61.24% <30.00%> (-0.05%) ⬇️
plugins-bucket-18 61.87% <30.00%> (-0.05%) ⬇️
plugins-bucket-19 61.25% <30.00%> (-0.05%) ⬇️
plugins-bucket-20 63.66% <30.00%> (-0.05%) ⬇️
plugins-bucket-4 58.26% <30.00%> (-0.05%) ⬇️
plugins-bullmq_cassandra_cookie 61.32% <30.00%> (-0.05%) ⬇️
plugins-cookie-parser_crypto_dd-trace-api 56.31% <30.00%> (-0.05%) ⬇️
plugins-fetch_fs_generic-pool 58.16% <30.00%> (-0.09%) ⬇️
plugins-google-cloud-pubsub_grpc_handlebars 64.09% <30.00%> (-0.04%) ⬇️
plugins-hapi_hono_ioredis 59.84% <30.00%> (-0.05%) ⬇️
plugins-knex_langgraph_ldapjs 55.02% <30.00%> (-0.05%) ⬇️
plugins-light-my-request_limitd-client_lodash 58.31% <30.00%> (-0.05%) ⬇️
plugins-mariadb_memcached_mercurius 61.23% <30.00%> (-0.05%) ⬇️
plugins-mongodb_mongodb-core_mongoose 59.20% <30.00%> (-0.05%) ⬇️
plugins-multer_mysql_mysql2 58.79% <30.00%> (-0.05%) ⬇️
plugins-nats_node-serialize_opensearch 60.34% <30.00%> (-0.05%) ⬇️
plugins-passport-http_pino_postgres 58.54% <30.00%> (-0.08%) ⬇️
plugins-process_pug_redis 57.34% <30.00%> (-0.05%) ⬇️
plugins-undici_url_valkey 57.97% <30.00%> (-0.05%) ⬇️
plugins-vm_winston_ws 59.54% <30.00%> (-0.05%) ⬇️
profiling 61.63% <30.00%> (-0.05%) ⬇️
serverless-aws-sdk-aws-sdk 55.09% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-base-inject-field 50.85% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-bedrockruntime 54.62% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-client 56.18% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-dynamodb 55.45% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-eventbridge 49.66% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-kinesis 59.03% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-lambda 57.19% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-s3 55.54% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-serverless-peer-service 59.29% <30.00%> (-0.05%) ⬇️
serverless-aws-sdk-sns 59.83% <30.00%> (-0.05%) ⬇️
serverless-aws-sdk-sqs 60.25% <30.00%> (-0.05%) ⬇️
serverless-aws-sdk-stepfunctions 55.38% <30.00%> (-0.04%) ⬇️
serverless-aws-sdk-util 51.38% <30.00%> (-0.04%) ⬇️
serverless-bucket-0 54.00% <30.00%> (-0.02%) ⬇️
serverless-bucket-1 58.80% <30.00%> (-0.05%) ⬇️
test-optimization-cucumber 71.37% <30.00%> (+0.11%) ⬆️
test-optimization-cypress 65.25% <30.00%> (+0.03%) ⬆️
test-optimization-jest 72.62% <30.00%> (-0.02%) ⬇️
test-optimization-mocha 72.18% <30.00%> (+0.01%) ⬆️
test-optimization-playwright-playwright-atr 59.74% <30.00%> (-0.23%) ⬇️
test-optimization-playwright-playwright-efd 60.08% <30.00%> (-0.04%) ⬇️
test-optimization-playwright-playwright-final-status 60.25% <30.00%> (-0.06%) ⬇️
test-optimization-playwright-playwright-impacted-tests 59.79% <30.00%> (+0.11%) ⬆️
test-optimization-playwright-playwright-reporting 61.29% <30.00%> (-0.09%) ⬇️
test-optimization-playwright-playwright-test-management 60.75% <30.00%> (-0.13%) ⬇️
test-optimization-playwright-playwright-test-span 59.97% <30.00%> (-0.10%) ⬇️
test-optimization-selenium 59.13% <30.00%> (-0.17%) ⬇️
test-optimization-testopt 57.71% <30.00%> (+0.05%) ⬆️
test-optimization-vitest 73.32% <30.00%> (+<0.01%) ⬆️
test-optimization-vitest-browser 58.96% <30.00%> (-0.04%) ⬇️
test-optimization-webdriverio 65.31% <30.00%> (+0.02%) ⬆️

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.

@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: 8ea014feb4

ℹ️ About Codex in GitHub

Codex has been enabled to automatically 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 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

this._tagger.tagToolDefinitions(span, toolDefinitions)
}
if (agent?.version != null) {
this._tagger.tagAgentVersion(span, agent.version)

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P2 Badge Drop ignored agent annotations before validating

When llmobs.annotate() is called on a non-agent span with agent.version, this call validates the version even though the span processor later drops _ml_obs.agent_version for every non-agent kind. A JavaScript caller such as llmobs.trace({ kind: 'workflow' }, () => llmobs.annotate({ agent: { version: 3 } })) therefore throws in the application path for a value that cannot be emitted; the already-commented trace/wrap path has the same root, and this annotate surface is a separate public entry point that needs the same kind gate or drop-before-validation behavior.

AGENTS.md reference: AGENTS.md:L222-L225

Useful? React with 👍 / 👎.

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.

1 participant