Skip to content

[codex] Add context-preserving invoke stream#21

Merged
senamakel merged 2 commits into
mainfrom
openhuman/invoke-stream-context
Jul 5, 2026
Merged

[codex] Add context-preserving invoke stream#21
senamakel merged 2 commits into
mainfrom
openhuman/invoke-stream-context

Conversation

@senamakel

Copy link
Copy Markdown
Member

Summary

Adds AgentHarness::invoke_stream_in_context, a caller-consumable streaming entry point that preserves a caller-supplied RunContext.

invoke_stream still constructs a fresh context for the simple case, but now delegates to the context-preserving method. This lets host integrations attach cancellation, events, stores, workspace metadata, or steering before driving the stream.

Validation

  • cargo fmt --check
  • timeout 180s cargo clippy --all-targets -- -D warnings
  • timeout 120s cargo test invoke_stream_in_context_preserves_caller_context

Full CI is left to GitHub runners.

@coderabbitai

coderabbitai Bot commented Jul 5, 2026

Copy link
Copy Markdown

Warning

Review limit reached

You’ve reached a temporary PR review limit under our Fair Usage Limits Policy.

Your recent review volume is higher than typical usage, so adaptive limits are currently applied.

Next review available in: 52 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details
⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 4d20e46b-91af-4191-8864-0d68e2ad56f8

📥 Commits

Reviewing files that changed from the base of the PR and between e72036d and a788b85.

📒 Files selected for processing (3)
  • src/harness/agent_loop/stream.rs
  • src/harness/agent_loop/test.rs
  • src/harness/events/mod.rs

Comment @coderabbitai help to get the list of available commands.

@senamakel
senamakel marked this pull request as ready for review July 5, 2026 03:30

@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

ctx.events.subscribe(Arc::new(ChannelListener { tx }));

P2 Badge Avoid leaking stream listeners on shared EventSink

When callers pass a RunContext that was built with a long-lived/shared EventSink via with_events, this one-shot channel listener is appended to that sink permanently; EventSink::subscribe stores listeners in its internal Vec and there is no unsubscribe path. After the returned stream completes or is dropped, the receiver is gone but the sender remains in the shared sink, so repeated invoke_stream_in_context runs leak listeners and make every later emit clone/send to dead channels.

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".

@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: a788b85500

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".

// clone, so their lifecycle events reach this stream too.
ctx.events.subscribe(Arc::new(ChannelListener { tx }));
let listener: Arc<dyn EventListener> = Arc::new(ChannelListener { tx });
ctx.events.subscribe(listener.clone());

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 Avoid starving runs on shared event sinks

When ctx.events is a long-lived/shared sink, this subscribes the stream listener before this run future has been polled. The unfold below uses a biased select that drains rx.recv() before polling run_fut, so events emitted by another active run or host component on the same sink can keep this stream yielding unrelated records and prevent this run from starting or completing. Consider isolating/filtering the listener to this run or ensuring the run future is polled even while the shared sink is busy.

Useful? React with 👍 / 👎.

@senamakel
senamakel merged commit 600470d into main Jul 5, 2026
3 checks passed
@senamakel
senamakel deleted the openhuman/invoke-stream-context branch July 5, 2026 03:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant