Skip to content

feat(mcp): Support explicit transformed time-series searches - #2017

Open
nsdeschenes wants to merge 2 commits into
feat/exp-1408-time-series-resultsfrom
feat/exp-1408-explicit-time-series
Open

nsdeschenes wants to merge 2 commits into
feat/exp-1408-time-series-resultsfrom
feat/exp-1408-explicit-time-series

Conversation

@nsdeschenes

@nsdeschenes nsdeschenes commented Nov 5, 2026 •

Copy link
Copy Markdown
Contributor

Expose ordered per-axis pipelines on trace, log, and metric searches for EXP-1408. Explicit searches run without an AI provider, and explicit settings survive translation and provider fallback. Embedded-agent and Seer routes normalize into the shared model; Explore links reproduce the applied axes, pipelines, and interval.

The three tools use the shared structured result envelope and declare ordered series[] and requestedInterval fields for explicit, transformed, or multi-axis results. Ordinary inferred single-axis results retain the existing timeSeries shape. Text output remains equivalent to the structured payload, and legacy search_events continues to return text.

Regression coverage checks SDK dispatch and output schemas, all three datasets, precedence, retries, response validation, URL reconstruction, and Seer outcome reporting after result validation.

@vercel

vercel Bot commented Nov 5, 2026 •

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
sentry-local Ready Ready Preview Nov 9, 2026 12:03pm UTC

Request Review

@linear-code

linear-code Bot commented Nov 5, 2026

Copy link
Copy Markdown

EXP-1408

@cursor cursor Bot left a comment •

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Stale Bugbot comment from a previous run.

@cursor cursor Bot left a comment •

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Stale Bugbot comment from a previous run.

@nsdeschenes
nsdeschenes added this pull request to stack #2049 November 6, 2026 15:19
@nsdeschenes nsdeschenes changed the title feat(mcp): support explicit transformed time-series searches feat(mcp): Support explicit transformed time-series searches Nov 6, 2026

@cursor cursor Bot left a comment •

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Stale Bugbot comment from a previous run.

@cursor cursor Bot left a comment •

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Stale Bugbot comment from a previous run.

@nsdeschenes
nsdeschenes marked this pull request as ready for review November 9, 2026 11:40
@github-actions github-actions Bot added the risk: medium PR risk score: medium label Nov 9, 2026
nsdeschenes and others added 2 commits November 9, 2026 08:01
Expose ordered pipelines for trace, log, and metric searches without an AI
provider. Preserve explicit settings through translation and fallback and
reconstruct the applied configuration in Explore links.

Refs EXP-1408

Co-Authored-By: GPT-6.1 Sol <noreply@openai.com>
Exclude Seer charts without a matching interval from inferred time-series
axes. Reject transformations on unbucketed charts instead of dropping them;
explicit caller overrides retain precedence.

Co-Authored-By: GPT-6.1 Sol <noreply@openai.com>

@cursor cursor Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Fix All in Cursor

❌ 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 d138436. Configure here.

),
transformations: agentTransformationsSchema,
}),
])

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Agent union silently drops extra axes

Medium Severity

The embedded agent still documents timeSeries.yAxis, while the new output schema tries a strict axes object first and then a non-strict legacy object. Zod keeps strict() across extend(), so a payload with both axes and yAxis fails the first branch and the second strips axes. normalizeTimeSeriesConfig then keeps only the legacy single series, dropping extra axes and per-axis transformations.

Fix in Cursor Fix in Web

Reviewed by Cursor Bugbot for commit d138436. Configure here.

This branch was successfully deployed

2 active deployments
Preview — d1384366 Deployed Nov 9, 2026 by vercel[bot]
Actions — d1384366 Deployed Nov 9, 2026 by nsdeschenes via eval #1742
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

risk: medium PR risk score: medium

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant