Loading ToolBench
Unofficial, community-built Model Context Protocol server that exposes the Wealthsimple Help Center (https://help.wealthsimple.com/hc/en-ca) to AI agents via the public Zendesk Help Center JSON API. Not affiliated with Wealthsimple Technologies Inc.
Strong tool definitions with clear naming, comprehensive descriptions, and well-structured schemas. All 6 tools follow verb_noun convention (search_, list_, get_, resolve_). Descriptions are detailed (150-250 chars) and explain WHEN to use each tool. Input schemas use Zod with proper type constraints (min/max lengths, enums, optional flags). Output schemas are documented with structured payloads. Key strength: discovery tools (list_categories, list_sections, list_articles) form a coherent taxonomy chain. Weakness: no error handling guidance in descriptions (e.g., what happens if article_id is invalid?), and no explicit mention of rate limits or pagination behavior in search results.
Fetch the full content of a single article by ID. Returns the article body in your choice of format (markdown, HTML, or plain text), plus metadata (title, URL, labels, section, locale, update timestamp). Use this to retrieve the canonical, detailed content for citations and comprehensive answers.
Return article summaries (no body) inside a single section. Use after `list_sections` to discover specific articles, then call `get_article` for full content.
List the top-level categories (Get Started, Move Money, Investing, Spending, File Taxes, Your Profile) of the Wealthsimple Help Center. Useful as the entry point for taxonomy-driven exploration before drilling into sections and articles.
List sections, optionally filtered to one category. Sections group related articles (e.g. "TFSA", "Crypto deposits"). Pass a `category_id` from `list_categories` to scope.
Resolve a public Wealthsimple help center URL (e.g. https://help.wealthsimple.com/hc/en-ca/articles/4404053510299-Open-a-TFSA) to full article content. Parses the numeric article ID from the URL, fetches the article, and returns it in your choice of format (markdown, HTML, or plain text). Useful when a user provides a help center link and you need the full content for citation or analysis.
No error recovery guidance in tool descriptions. When search_help_center returns 0 results or get_article fails with invalid ID, descriptions don't tell the LLM what to do next (e.g., 'Try a broader query' or 'Verify the article_id from list_articles').
search_help_center description mentions 'ranked list' but doesn't explain ranking criteria (relevance score? recency?). LLMs may assume results are exhaustive when they're actually truncated at limit=10 by default.
resolve_help_url accepts arbitrary URLs but doesn't document what happens if the URL is malformed, from a different domain, or points to a non-article page (e.g., category page). No validation error guidance.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 59 | 2025-06-18+ | v2 |
Full-text search across every public Wealthsimple Help Center article. Returns a ranked list of articles with titles, URLs, snippets, labels, and section IDs. Use this whenever the user asks anything about Wealthsimple products (Trade, Invest, Cash, Crypto, Tax, registered accounts like TFSA / RRSP / FHSA, transfers, deposits, etc.) — the help center is the canonical source of truth.
list_articles and list_sections accept optional limit parameters but don't document default behavior or explain when to use pagination. LLMs may not realize they need to call these tools multiple times for large result sets.
Output schemas for list_* tools don't include a 'total_count' or 'has_more' field, making it unclear to the LLM whether results are complete or truncated.