datasetsSearch the Providence Open Data catalog of open datasets by keyword. Returns each dataset's resource_id, name, description, category and update date — pass the resource_id to query/metadata.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Max datasets (1-100, default 20). | |
| query | No | Keyword to search dataset titles/descriptions (e.g. "budget", "crime", "health"). | |
| offset | No | Pagination offset. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare read-only, non-destructive, idempotent. Description adds value by listing returned fields and usage hint for resource_id, going beyond annotation basics.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two efficient sentences, front-loaded with action and results. Every word adds value, no repetition or filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Combined with annotations and full schema, description fully covers purpose, inputs, output hints, and next step. No gaps for a search tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema covers all parameters with 100% description coverage, so description need not add much. Baseline 3 is appropriate; description doesn't enhance parameter meaning.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool searches a specific open data catalog by keyword, with precise verb+resource+scope. It distinguishes from 25+ sibling tools as a focused dataset search.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Describes return fields and next step (pass resource_id to query/metadata). Missing explicit when-not-to-use or alternative tools, but context is clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools have distinct purposes, but several clusters overlap: ask_pipeworx vs ask_pipeworx_beta are currently functionally identical, polymarket_arbitrage / polymarket_edges / polymarket_kalshi_spread all hunt mispricings via different mechanisms, ai_visibility_check is wrapped by scan_competitor_ai_presence, and discover_tools vs suggest_questions both serve tool discovery. The rich descriptions mitigate but do not eliminate misselection risk.
Names are all snake_case and follow recognizable conventions: verb_noun for actions (compare_entities, resolve_entity, validate_claim), domain-prefixed families (polymarket_*, pipeworx_*, recent_*, ask_pipeworx_*), and a few bare verbs (remember, recall, query). Minor deviations like bet_research (noun_verb) and noun-only names (datasets, metadata) break the pattern, but the overall scheme is predictable.
At 34 tools, this exceeds the 25+ threshold for 'too many' and bundles several distinct domains — general data querying, prediction markets, AI visibility, memory, subscriptions, open data, and npm auditing — into one server. The breadth is defensible for a data platform, but the agent-facing surface is sprawling and would benefit from splitting into focused servers.
The core data workflow is well covered: discover (discover_tools, suggest_questions), resolve (resolve_entity), query (ask_pipeworx), ground (ask_pipeworx_grounded, validate_claim), research (deep_research), compare (compare_entities), profile (entity_profile), and changes (recent_changes). Prediction markets, memory, and subscriptions each have full lifecycles. The main gap is no tool for fetching returned pipeworx:// citation URIs directly, plus a few soft-failing sources.