Loading ToolBench

Reading the current corpus…

agentprism-workflows MCP server · F · ToolBench

agentprism-workflows

github.com/agentprism/agentprism-workflows

A monorepo containing multiple MCP servers and agents for workflow automation, including ACP agents, MCP server implementations, and structured output tools

2025-06-18+STREAMABLE_HTTP1 tools

38/100

Definition Quality27

Protocol Readiness65

SupportabilityNot measured-

Spec posture

Inferred effective spec 2025-06-18+·current baseline 2026-07-28

Static source inference · medium confidence · evidence: structured output, Streamable HTTP

No deprecated protocol patterns detected

Summary

Single tool 'StructuredOutput' with critical definition gaps. Input schema is declared as 'Dynamic schema based on registration', no actual schema visible in source code. Description is present but overly prescriptive (tells agent WHEN to call, not WHAT it does). No parameter documentation. Tool appears to be a meta-tool for formatting responses rather than a functional capability. Cannot verify schema structure, types, or constraints from provided code.

Tools (1)

StructuredOutputread onlysource verified27/100

Use this tool to return your final response in the requested structured format. IMPORTANT: - You MUST call this tool exactly once at the end of your response - The input must be valid JSON matching the required schema - Complete all necessary research and tool calls BEFORE calling this tool - This tool provides your final answer - no further actions are taken after calling it

Issues (5)

CRITICALtool ↗

Input schema is not defined, declared as 'Dynamic schema based on registration' with no actual JSON Schema visible.

StructuredOutput

HIGHtool ↗

Tool name 'StructuredOutput' does not follow verb_noun convention. Should be 'format_output', 'return_structured_output', or similar action verb. Current name is a noun phrase that obscures intent.

StructuredOutput

HIGHtool-description ↗

Description is prescriptive (tells agent WHEN/HOW to call) rather than descriptive (WHAT it does, WHY to use it). Reads as a constraint, not a tool definition. LLMs need to understand the tool's purpose, not instructions for its use.

StructuredOutput

HIGHtool-description ↗

No parameter descriptions. Input is a generic object with no field-level documentation. LLMs cannot infer what fields are required, what types they accept, or what constraints apply.

StructuredOutput

Recommendations

Define explicit JSON Schema for input with typed fields, constraints, and descriptions. Replace 'Dynamic schema based on registration' with actual schema properties.
Rename tool to action verb form: 'format_response', 'return_structured_output', or 'finalize_response' to clarify intent.
Rewrite description to explain WHAT the tool does and WHEN to use it, not HOW. Example: 'Formats your final response as structured JSON matching the requested schema. Call this once after completing all research and tool calls.'
Add per-field documentation for all input parameters. Specify required vs optional, type, format, and constraints for each field.
Implement actual domain tools (e.g., workflow execution, data retrieval, state management) rather than relying on a single meta-tool for output formatting.

Score history

Overall score trend

First recorded score · v2 rubric

38/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	F	38	2025-06-18+	v2