Loading ToolBench
Alpha MCP server for local LLM memory, knowledge, state, routing, testing, and orchestration.
Server has 8 tools with basic schemas and descriptions, but significant gaps limit production readiness. All tools have descriptions (10-60 chars, mostly short), and input schemas are present with type definitions. However, descriptions lack LLM-optimized detail (pattern:tool-description baseline 194 chars; these average ~35 chars). Parameter descriptions are minimal or absent. Output schemas are undocumented. Error handling is not visible in tool definitions. No tool annotations (readOnlyHint/destructiveHint) despite clear risk classifications. Naming follows verb_noun convention (hb_api_*, hb_auto_*) but lacks clarity on when to use similar tools (e.g., hb_api_probe vs hb_api_discover). No evidence of pagination, result limits, or chaining IDs in responses.
Auto-detect API schema from a base URL
Export probe results as Markdown or JSON
List previous probe results
Probe a URL using all strategies (OpenAPI, wordlist, pattern, HATEOAS)
List available local automation chain definitions
Get the recorded local automation run result
Queue a local automation chain plan without executing external backends
Descriptions are too short (avg 35 chars vs baseline 194 chars). Missing WHEN to use each tool, prerequisites, and return value guidance. LLMs cannot reliably select between hb_api_probe and hb_api_discover without richer context.
Parameter descriptions are missing or minimal. 'strategies' array in hb_api_probe lacks explanation of what each strategy does (openapi vs wordlist vs pattern vs hateoas). 'input' in hb_auto_run has no format guidance. LLMs cannot infer parameter semantics from names alone.
No output schemas documented. Callers cannot know what fields to expect from probe results, automation run results, or history. This breaks tool chaining and forces LLMs to guess at response structure.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | <=2025-11-25 | v2 |
Check local automation run status
No tool annotations (readOnlyHint/destructiveHint/idempotentHint) despite clear risk classifications. hb_auto_run is marked WRITE but lacks destructiveHint. Clients cannot determine which tools are safe to retry or require confirmation.
No pagination or result limits visible in tool definitions. hb_api_history accepts 'limit' but no documentation of max value, default behavior, or cursor/offset support. Large result sets risk context window exhaustion.