Loading ToolBench
Advanced Custom Fields Model Context Protocol server integration for Angie AI. Provides comprehensive tools for managing ACF field groups, fields, custom post types, and taxonomies in WordPress.
Server defines 4 tools with Zod schemas and descriptions. Naming follows verb_noun pattern (acf-field, acf-field-group, acf-post-type, acf-taxonomy). Descriptions are present (194 - 250 chars, within baseline 34 - 392 range) and explain WHAT the tool does and WHAT it returns. However, parameter descriptions lack actionable constraints (no ranges, patterns, or format specs). Schema objects (ACFFieldSchema, ACFFieldGroupSchema, etc.) are referenced but not visible in source, cannot verify their completeness. Error handling returns generic messages without recovery guidance. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite 4 WRITE-risk tools. Missing output schema documentation.
This tool manages ACF field operations, allowing you to create, update, retrieve, and delete individual fields within field groups. **What does it do?** - Retrieves individual fields or lists of all fields - Creates new fields within existing field groups - Updates field configurations and properties - Deletes fields from field groups - Handles all field types with their specific properties **What it returns?** Returns JSON data for field configurations, success confirmations for create/update operations, and proper error messages for validation issues.
This tool manages ACF field groups, which are containers that hold related fields and define where they appear in WordPress. **What does it do?** - Manages field group creation, updates, and deletion - Configures field group display locations and rules - Provides ACF plugin status checking - Generates intelligent field structure suggestions for a given post type - Handles field group properties like positioning, styling, and visibility **What it returns?** Returns JSON data for field group configurations, success confirmations for create/update operations, ACF status information, and structured field suggestions.
This tool manages custom post types in WordPress through ACF, allowing you to create specialized content types beyond the default posts and pages. **What does it do?** - Creates custom post types with full configuration options - Retrieves existing post type(s) and their configurations - Manages post type properties, labels, and capabilities - Updates existing post type settings and permissions - Handles post type deletion and cleanup - Configures post type behavior (hierarchical, public, searchable, etc.) **What it returns?** Returns JSON data for post type configurations, success confirmations for create/update operations, and detailed error messages for configuration issues.
Schema objects (ACFFieldSchema, ACFFieldGroupSchema, ACFPostTypeSchema, ACFTaxonomySchema) are imported but not visible in source code. Cannot verify field types, constraints, or completeness. Scoring schema as 50 (inferred, not verified).
All 4 tools are marked WRITE risk but lack tool annotations (destructiveHint, idempotentHint). LLMs cannot determine if operations are safe to retry or require confirmation. No dry-run or confirmation_before_execute pattern implemented.
Parameter descriptions lack actionable constraints. E.g., 'action' enum is documented but no guidance on which parameters are required for each action value. 'field' and 'fieldGroup' objects lack schema details (required fields, types, nested structure).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | 2026-07-28+ | v2 |
This tool manages custom taxonomies in WordPress through ACF, enabling you to create, retrieve, update and delete custom taxonomies
Error handling returns generic 'Error performing [operation]' messages without recovery guidance. No categorization of errors as retryable, user-fixable, or fatal. LLM cannot determine next action.
Output schema not documented. Tool descriptions state 'Returns JSON data' but do not specify field names, types, or structure. LLM cannot plan downstream tool calls or extract required data.