get-dataset-itemsRead result rows from a caller-owned Apify dataset, with bounded pagination and optional field selection.
| Name | Required | Description | Default |
|---|---|---|---|
| desc | No | ||
| omit | No | ||
| clean | No | ||
| limit | No | ||
| fields | No | ||
| offset | No | ||
| flatten | No | ||
| datasetId | Yes | Dataset ID or username~dataset-name. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover read-only, idempotent, and non-destructive behavior, so the bar is lower. The description adds useful context about caller ownership and bounded pagination, but it does not disclose concrete behavioral details such as pagination limits, defaults, or how field selection behaves.
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?
The description is a single front-loaded sentence with no filler. Every phrase adds meaning: the operation, the ownership constraint, pagination behavior, and field-selection capability.
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?
With 8 parameters, no output schema, and minimal parameter descriptions, the description alone is not enough for an agent to invoke the tool correctly. It fails to explain return shape, pagination bounds, field selection syntax, flatten/omit/clean semantics, or default behavior.
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 description coverage is only 13%, so the description carries a heavy burden for explaining parameters. It vaguely covers pagination (limit/offset) and field selection (fields), but completely omits desc, omit, clean, and flatten, leaving seven of eight parameters effectively undocumented.
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 operation ('Read result rows') and the resource ('caller-owned Apify dataset'), which distinguishes it from sibling tools like get-key-value-store-record and get-actor-run. The phrase 'bounded pagination and optional field selection' adds concrete scope beyond the title.
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?
The description conveys a clear context: this tool is for reading result rows from a caller-owned dataset, which implies it is not for other dataset owners or other data stores. It does not name alternative tools explicitly, but there are no close dataset siblings, so the context is sufficient.
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 domain tools are well-scoped with explicit cross-references (e.g., FL DBPR vs. STR registry, Texas vs. multistate childcare). However, the evidence-gateway overlaps with EPA, U.S. property, and other specialized tools by describing similar intake categories, creating ambiguity about when to use the router versus the domain-specific tool.
The specialized tools consistently use the civicdataforge-- prefix with descriptive noun phrases, while the generic actor tools use imperative verb_noun style. The naming is readable and predictable within each subgroup, but the mixed conventions and the awkward doubled prefix in civicdataforge--civicdataforge-evidence-gateway prevent full consistency.
Fourteen tools is reasonable for a broad government-evidence server covering many data domains plus an async run lifecycle. The count is not excessive, though the gateway and several overlapping domain-specific tools add some redundancy.
The tool set covers a wide range of evidence domains and provides complete async workflow coverage: launch queries, check run status, fetch dataset items, read KVS records, and abort runs. Minor gaps remain, such as no explicit way to enumerate supported jurisdictions or sources, and the gateway's broad categories are underspecified.