Extend MCP

Get an evaluation run

get_evaluation_run

Read-onlyIdempotent

Get an evaluation run's status and accuracy metrics — this is the ONLY tool that returns them, not a get-batch tool (evaluations group). An evaluation is an async job: a fresh run reports PROCESSING, so pass wait: true to block until it finishes instead of polling in a loop. Metrics by resource type: extractors { accuracy, fieldMetrics per field path — each field has countExpected/countAccurate, and accuracy is aggregated across items, not per-item }; classifiers { accuracy, classificationMetrics with precision/recall/f1 per type }; splitters { precision, recall, f1, split counts }. Terminal statuses: PROCESSED, FAILED, CANCELLED. Follow any llmContext guidance included in results.

Input Schema

TableJSON Schema

Name	Required	Description
`wait`	No	Poll until the evaluation run reaches a terminal status (within the wait budget).
`environment`	Yes	"TEST" = the Test (development) environment, "PRODUCTION" = live. Must match a granted target from get_me (an API key pins one environment).
`waitSeconds`	No	Max seconds to block waiting on the run (clamped to the server wait budget).
`workspaceId`	Yes	Target workspace (ws_...). Must be a granted workspace — get_me lists the accepted values.
`evaluationRunId`	Yes	The bpr_... ID returned by run_evaluation.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`id`	Yes
`entity`	No
`status`	Yes
`metrics`	No
`llmContext`	No
`entityVersion`	No
`evaluationSetId`	No

TDQS

A4.8/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/non-destructive, and the description adds substantial behavior: async processing with PROCESSING status, terminal statuses, wait semantics, and metric aggregation rules. Also notes llmContext guidance should be followed. No contradictions.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Dense but structured description, front-loaded with purpose and exclusivity. The metric breakdown by resource type is long but useful for the agent to know what to expect; no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Complete for a read-only retrieval tool: describes statuses, metrics by resource, wait behavior, terminal states, and post-processing guidance. Output schema further covers return shape.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so baseline is 3. The description adds meaning beyond the schema by explaining wait:true as a blocking alternative to polling, and clarifies metrics are aggregated across items rather than per-item. This enriches the parameter's purpose, though most param details remain in the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb ('Get') and resource ('evaluation run's status and accuracy metrics'), and explicitly distinguishes itself from sibling get-batch tools in the evaluations group, saying it is the ONLY tool that returns these. This makes selection unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly tells the agent when to use it (to obtain status/metrics), how to handle async runs (pass wait:true instead of polling), and references terminal statuses and llmContext guidance. Also implies this is for evaluation runs, vs batch tools for other groups.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Extend MCP tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4/5.0

Disambiguation5/5

Each tool targets a distinct resource+action combination, and the descriptions actively disambiguate potential overlaps (e.g., extract_data vs parse_document, detect_form_fields vs edit_pdf, get_file vs get_file_upload). The consistent verb_noun prefix pattern makes the semantic boundary of every tool immediately recognizable.

Naming Consistency4/5

The dominant verb_noun pattern is highly consistent across all nine domains (list_*, get_*, create_*, update_*, delete_*, run_*, get_*_run, get_*_batch, publish_*_version). Minor deviations exist: deploy_workflow_version vs publish_*_version for the same freeze-a-draft concept, and get_form_detection_run doesn't mirror its detect_form_fields counterpart.

Tool Count2/5

86 tools is a very heavy agent-facing surface, well past the 25+ threshold. The count is inflated by the near-identical 13-tool lifecycle repeated across extract, classify, and split (each with list/get/create/update/publish/runs/batches/versions), and while each tool has a distinct purpose, the sheer volume makes selection harder.

Completeness4/5

Core lifecycles are thoroughly covered: create → update → publish → run (single and batch) → poll → cancel → delete-run → list runs/versions. Notable gaps include no delete tool for extractors, classifiers, splitters, workflows, or evaluation sets, and edit/form-detection runs have no list endpoint (documented workaround: keep run IDs). These are hygenic gaps that don't block primary workflows.

Resources

Need Help?