Repository navigation
Conversation
Contributor
size-limit report 📦Compared: f610bbf (PR test merge using develop at f1e8e61) Against: ec88aef (develop) · base run #64752
|
Contributor
Optional E2E testsSome optional checks were not successful. They do not block the merge. 6 not successful, 47 successful. See the run.
Successful checks
|
mydea
force-pushed
the
feat/pi-ai-integration-default
branch
from
October 30, 2026 12:53
7f58fd6 to
91831f6
Compare
mydea
force-pushed
the
feat/pi-ai-integration-default
branch
from
October 30, 2026 12:54
91831f6 to
c5e394f
Compare
mydea
marked this pull request as ready for review
October 30, 2026 13:21
mydea
requested review from
andreiborza,
chargome,
isaacs,
logaretm,
msonnb and
nicohrubec
and removed request for
a team
October 30, 2026 13:21
mydea
changed the title
chargome
reviewed
Nov 3, 2026
| langChainIntegration(), | ||
| langGraphIntegration(), | ||
| mastraIntegration(), | ||
| piAiIntegration(), |
Member
There was a problem hiding this comment.
Turning this on by default means one pi-ai request turns off the OpenAI / Anthropic / Google GenAI / Workers AI integrations for the whole process. So an app that uses pi-ai and also e.g. calls openai directly loses spans for its direct calls.
Could we make the skip apply only to requests pi-ai sends? The provider integrations already check the skip on every request, and the pi-ai chat span is active while pi-ai's request goes out. So they could skip only when they're inside a pi-ai span, and leave other calls alone.
Comment on lines
+55
to
+65
| consoleSandbox(() => { | ||
| // eslint-disable-next-line no-console | ||
| console.warn( | ||
| '[Sentry] Flue model calls are not being traced: `@earendil-works/pi-ai` was not instrumented, so ' + | ||
| 'Flue agents record agent and tool spans but no `gen_ai.chat` spans. Model calls are traced by the ' + | ||
| "pi-ai integration, which needs Sentry's module hook or bundler plugin. On Node, load Sentry with " + | ||
| '`--import` (for example `node --import ./instrument.mjs app.mjs`). On Cloudflare, add ' + | ||
| '`sentryCloudflareVitePlugin()`. On Bun, build with `@sentry/bun/plugin`.', | ||
| ); | ||
| }); | ||
| } |
Member
There was a problem hiding this comment.
l: We could detect the runtime and just print the according warning
mydea
added a commit
that referenced
this pull request
Nov 3, 2026
…ibraries (#25947) **Why:** The `gen-ai-libraries (latest)` variant installs `together-ai@latest`, which is now 1.0.0. That release moved `/v1` from the SDK's default base URL into the request paths, so with our `https://openrouter.ai/api/v1` base URL it requests `/api/v1/v1/chat/completions` and both Together cases fail with a 500 (e.g. on #25940). Against a local server, 0.54.0 hits `/api/v1/chat/completions` with the `/v1` base and 1.0.0 hits `/api/v1/v1/chat/completions`; with the base URL minus `/v1`, 1.0.0 hits the right path. The default variant keeps 0.54.0, so the client picks the base URL from `together-ai/version`, which both versions export. Verified against a local server for both versions, and the import and expression type-check against both. **How to review:** One line in `src/libraries.ts`. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Enables `piAiIntegration` by default. Flue and pi-durable send their model calls through pi-ai, so they no longer open chat spans of their own: Flue drops its turn spans and hands the conversation id to the pi-ai span through the scope, and pi-durable drops its `models` proxy. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Flue's model calls are now traced by the pi-ai integration, which needs the module hook or bundler plugin, while `instrument()` does not. Without them a Flue app would silently lose its `gen_ai.chat` spans; it now gets a warning on its first model call. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
mydea
force-pushed
the
feat/pi-ai-integration-default
branch
from
November 3, 2026 12:09
c5e394f to
52faa76
Compare
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Want reviews to match your repository better? Bugbot Learning can learn team-specific rules from PR activity. A team admin can enable Learning in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 52faa76. Configure here.
| key: FLUE_INSTRUMENTATION_KEY, | ||
|
|
||
| interceptor: async (operation, ctx, next) => { | ||
| skipProviders(); |
There was a problem hiding this comment.
Flue options no longer cover chats
Medium Severity
createFlueInstrumentation({ recordInputs, recordOutputs }) no longer controls model I/O on gen_ai.chat spans. Those spans now come from piAiIntegration, which only reads its own options and dataCollection.genAI. An existing Flue opt-out no longer suppresses prompts and completions, and an existing opt-in no longer records them. The JSDoc still says message content is gated on these Flue options.
Additional Locations (1)
Triggered by project rule: PR Review Guidelines for Cursor Bot
Reviewed by Cursor Bugbot for commit 52faa76. Configure here.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Why: Flue and pi-durable both send their model calls through pi-ai, and each had its own chat-span code (Flue's turn observations, pi-durable's
modelsproxy). WithpiAiIntegration(#25937) enabled by default, pi-ai becomes the single source ofgen_ai.chatspans, and ~1.1k lines of duplicated span code go away. Flue now hands its conversation id to the pi-ai span through the scope.Stack: #25932 → #25937 → this. #25935 must be merged before #25937 and this: without it, instrumenting pi-ai makes the Cloudflare Flue worker crash at startup. Supersedes #25869.
How to review: The first commit is the switch. The second adds a one-time warning when Flue runs but pi-ai was not instrumented; it's separate so it can be dropped or split off.
Unsure about:
auto.ai.flue/auto.ai.pi_durabletoauto.ai.pi_ai.instrument(createFlueInstrumentation())keeps agent and tool spans but loses chat spans, hence the warning. Setups following our docs are unaffected.flue.turn.purpose(compaction vs. agent turn) is dropped, since pi-ai cannot see it. A possible follow-up is to record compaction the way feat(attributes): Add gen_ai conversation compaction conventions sentry-conventions#723 proposes, once it merges: Flue emitscompaction_start(with amanual/threshold/overflowreason) andcompactionobservations, so it could open agen_ai.compactspan with the summarizinggen_ai.chatnested under it.modelsobject created with pi-ai'screateModels().Release note:
🤖 Generated with Claude Code