Repository navigation
fix: restore agent traces navigation icon - #628
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
@codex review |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 66ce073a77
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".
| for run in run_store.list_runs(agent["agent_id"]): | ||
| trace = trace_store.get_trace_for_run(run["run_id"]) |
There was a problem hiding this comment.
In the production Render configuration, protocol_runs is explicitly stored in the diskless SQLite database and reset on every deploy (render.yaml:107-115), while trace_store uses durable AGENT_RUNS_DATABASE_URL Postgres (domain/traces/repository.py:359-375). Starting this projection exclusively from run_store.list_runs() therefore makes every persisted trace from a previous deploy disappear from the user's “Agents & traces” section even though it remains available in the global trace view. Merge traces queried by agent ID into the projection rather than requiring a surviving local protocol-run row.
Useful? React with 👍 / 👎.
| for agent in agent_store.list_agents(owner_user_id=user_id): | ||
| runs = [] | ||
| for run in run_store.list_runs(agent["agent_id"]): | ||
| trace = trace_store.get_trace_for_run(run["run_id"]) |
There was a problem hiding this comment.
For an owner with a long run history, this performs an unbounded list_runs() for every agent and then a separate trace-store query for every returned run. Since the trace store is normally remote Postgres, opening this tab can generate hundreds or thousands of sequential round trips and an arbitrarily large response, eventually timing out the admin request. Add a bounded/paginated contract and fetch the matching traces in batches.
Useful? React with 👍 / 👎.
|
Addressed the review findings in 518b091: user activity now reads paginated durable traces by agent IDs, with local protocol runs used only for enrichment. Added coverage for the batch query contract. Local targeted tests pass. |
This branch was successfully deployed
1 participant
The Agent traces navigation item was updated to use a distinct document icon, but the admin page did not define that symbol in its local SVG sprite, so the icon rendered blank.
This PR adds the missing
icon-file-textsymbol and keeps it separate from the Activity waveform icon.Validation:
/adminHTML includes the symbol and the trace navigation reference.main.