Loading ToolBench

Reading the current corpus…

egograph MCP server · F · ToolBench

egograph

Personal data warehouse MCP server providing access to Spotify, GitHub, browser history, YouTube, and Google Health data via tools. Includes REST API and MCP endpoints.

2026-07-28+STREAMABLE_HTTP9 tools

45/100

Definition Quality42

Protocol Readiness65

Supportability38

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence · evidence: Streamable HTTP

Current-spec patterns detected

Summary

Server exposes 9 read-only tools via HTTP. Tool definitions are partially visible: 2 tools (get_daily_timeline, data_query) lack visible schemas in source; 7 tools have schemas inferred from test file. Descriptions are present but generic (10-60 chars). Parameter descriptions exist but lack format constraints, ranges, or validation guidance. No error handling documentation. No tool annotations. Schemas lack output documentation. This is typical of early-stage data access servers.

Tools (9)

activity_statsread onlyauth50/100

Get activity statistics with date range and granularity

data_queryread onlyauth37/100

Execute raw SQL queries (SELECT only) against the data warehouse

get_daily_timelineread onlyauth35/100

Retrieve daily timeline data with compact serialization

listening_statsread onlyauthsource verified57/100

Get listening statistics with date range and granularity

repositoriesread onlyauth45/100

Get repository list

top_channelsread onlyauthsource verified60/100

Get top channels with date range and limit

top_tracksread onlyauth50/100

Get top tracks with date range and limit

watch_historyread only

Issues (5)

CRITICALtool ↗

Two tools (get_daily_timeline, data_query) have no visible input schemas in source code. Cannot verify parameter types, constraints, or LLM-usable structure.

get_daily_timelinedata_query

HIGHtool-description ↗

Descriptions are too short (10-60 chars) and lack context. 'Retrieve daily timeline data with compact serialization' does not explain WHEN to use it, WHAT it returns, or HOW it differs from other tools. LLMs cannot reliably select the right tool.

get_daily_timelinerepositoriesdata_query

HIGHconstrained-input ↗

Parameter descriptions lack format constraints and validation rules. E.g., 'start_date' and 'end_date' have no format (ISO 8601?), no range (past 1 year only?), no example. LLMs will guess and pass invalid dates.

watch_historywatching_statslistening_statstop_trackstop_channelsactivity_stats

Recommendations

Add explicit input schemas to get_daily_timeline and data_query in mcp_server.py. Include type, description, and constraints for every parameter.
Expand tool descriptions to 100-200 chars. Example: 'watch_history' → 'Retrieve user watch history within a date range. Returns video titles, timestamps, and duration. Use this to analyze viewing patterns or find specific videos watched on a date.'
Add format and range constraints to date parameters: 'start_date (ISO 8601, e.g. 2026-01-01, must be ≤ 1 year ago)' and 'end_date (ISO 8601, must be ≥ start_date and ≤ today)'.
Document output schemas for all tools. Specify field names, types, and what each field represents. E.g., watch_history returns [{title: string, timestamp: ISO8601, duration_seconds: integer}, ...].
Add error handling guidance: 'If no results found, try expanding the date range or reducing the limit. If date format is invalid, use ISO 8601 (YYYY-MM-DD).'

Score history

Overall score trend

First recorded score · v2 rubric

45/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	F	45	2026-07-28+	v2