Skip to main content
Glama
Sign In

Assessment Intelligence

by ai.saifs
OverviewDiscussionsAdmin

Server Details

Analyse tests for item quality, true ability, reliability and learner mastery.

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
OAuth
Works in Glama
Last Tested
Transport
Streamable HTTP · MCP 2025-06-18
URL
Repository
Saify-Technologies/saifs-ai-mcps-docs
GitHub Stars
0
ChatGPT
–
Claude
–

TDQS

C2.9/5.0

Scored across 2 tools

Disambiguation4/5

The two tools are mostly separable: one computes analytics in-memory (up to 5,000 rows) and the other returns a Python script for larger sandbox files. However, the primary tool's name 'assessment_intelligence' is a monolithic label covering item analysis, alpha, Rasch, BKT, and adaptive item selection, so an agent cannot tell from the name which capability it is invoking or how parameters differ per mode.

Naming Consistency4/5

Both names share a consistent server prefix ('assessment_intelligence'), with the second appending a clear '_get_engine' suffix. The inconsistency is minor: the first tool has no verb at all while the second uses a get_ pattern, so the convention is not uniformly verb_noun.

Tool Count3/5

Two tools for a rich psychometric/analytics domain is thin—nearly the entire surface is one mega-tool. The second tool only exists as a scaling escape hatch, which is reasonable but leaves the set borderline under-scoped.

Completeness3/5

The main tool covers many analyses but collapses them into a single opaque interface, and there are no tools to list available response files, retrieve or persist prior results, or check engine status. The get_engine tool mitigates the row-limit gap, but notable lifecycle operations are missing.

Publisher details

Operator
Saifs.ai · Publisher source
Operator website
https://saifs.ai/ · Publisher source
Vendor relationship
Not applicable
Documentation
Not applicable
Trust center
Not applicable
Restrictions
Not applicable

Available Tools

2 tools
assessment_intelligenceAssessment IntelligenceC
Read-only
Inspect

Item analysis, Cronbach's alpha, Rasch ability estimates with standard errors, Bayesian knowledge tracing, an adaptive next item, and support flags. Never ranks learners. Up to 5,000 rows.

Inputs: item_bank, responses, asked_items, focus_learner, +3 more. Returns: object

assessment_intelligence_get_engineAssessment Intelligence — Get Engine ScriptB
Read-only
Inspect

Returns a Python script that runs the same engine on a larger response file already in the sandbox.

Inputs: params, data_kind, data_path. Returns: object

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • whoqol-toolkit

      LaBiOmicS
      A
      license
      Not graded
      quality
      C
      maintenance
      Enables scoring and statistical analysis of WHOQOL quality-of-life questionnaires using official WHO rules, including data quality checks, psychometrics, group comparisons, APA-ready tables and figures, and draft methods/results text.
      MIT

    • Psychometrics MCP

      huacheng1985
      A
      license
      B
      quality
      C
      maintenance
      Enables AI hosts to perform reproducible psychometric and statistical analyses through constrained, schema-validated tools such as item analysis, Rasch modeling, descriptive statistics, and correlation matrices, without exposing arbitrary R or shell execution.
      7
      MIT

    • flakyguard

      CloudCorpRecords
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables analysis of per-test pass/fail outcomes across CI runs to compute evidence-based flakiness rates, classify tests as stable, flaky, or insufficient data, and produce a ranked quarantine list with a verdict. All computation happens locally from supplied run data, with invalid input rejected before any partial analysis.
      308 npm
      MIT
    Try in Browser
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    Glama MCP Gateway

    Add one secure layer between your agents and this server.

    Resources