Repository navigation
Conversation
Experiment, not meant to merge. The feedback page is built around people using the widget, but an MCP server's feedback is mostly agents that got stuck, and for those the trace preview is always empty because every tool call is its own trace. Agent reports get an Agent Session section instead: the tool calls that user made before the report, pulled from mcp.server spans, with the arguments, what the tool returned and the logs on each call. The failed call the report is about opens by default. Tool Health puts the reported tool's call volume and failure rate next to it. The list gets a reports-by-tool table with call counts and failure rates, impact filters, and rows titled by the task instead of the reporter name. Not behind a flag. Agent reports are detected by reporter name in the list since the issue doesn't carry the feedback source. Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Contributor
📊 Type Coverage Diff
🔍 5 new type safety issues introduced
Non-null assertions (
This is informational only and does not block the PR. |
This branch was successfully 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Experiment only, not intended to be merged. Opening as a draft to share the idea.
The feedback page is built around people using the widget, but feedback on an MCP server mostly comes from agents that got stuck. For those the trace preview is always empty since every tool call is its own trace, and the structure agents send (task, tool, impact) is buried in the tags.
For agent reports this swaps the trace preview for an Agent Session section that lists the tool calls that user made before the report, with the arguments, what the tool returned and the logs for each call. The failed call the report is about opens by default. A Tool Health section shows the reported tool's call volume and failure rate, and the list gets a reports-by-tool table with impact filters.
Not behind a flag and no tests. Agent reports are detected by reporter name in the list because the issue doesn't carry the feedback source.