Wayback Machine

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already indicate readOnly, openWorld, and idempotent behavior. The description adds worthwhile behavioral context beyond those annotations: it returns the raw-bytes URL, strips HTML and caps text at 30k chars, and falls back to Common Crawl when the Wayback Machine has no capture. These details help an agent predict what it will receive.

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?

Every sentence earns its place: the first defines the core function, the second lists return values, the third covers fallback behavior, and the fourth gives a concrete usage cue. It is compact, readable, and front-loaded with the most important information.

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 no output schema, the description compensates well by listing the returned fields: snapshot URL, exact timestamp, raw-bytes URL, and optional page text. It also handles the no-capture case via Common Crawl fallback. Minor gaps like exact response structure or behavior when no archive exists anywhere are not specified, but the description is sufficient for correct selection and invocation.

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 coverage is 100%, so the baseline is 3, but the description adds meaningful parameter-related semantics: it explains that 'timestamp' selects the closest capture to a date, and that 'fetch_text' returns HTML-stripped text capped at 30k characters. This goes beyond the schema's terse parameter descriptions.

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 identifies the resource (archived copy of a URL from the Wayback Machine), the action (returns the closest snapshot with timestamp, raw-bytes URL, and optional page text), and its fallback behavior. It also differentiates itself from the sibling capture_history by emphasizing 'closest' single snapshot rather than a list of captures.

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 gives an explicit use case: 'Use when a page is dead, changed, or walled and you need what it said.' This clearly communicates when to use the tool, though it does not explicitly mention when not to use it or compare it to the capture_history sibling.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, openWorldHint, and idempotentHint, so the safety profile is well covered. The description adds meaningful behavioral depth beyond annotations: CDX-index semantics, the collapse dedupe (one capture per period), the full namespace of matching modes (URL, prefix, host, domain), and pagination via resume_key. No contradictions.

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?

Three sentences with zero filler: the first fixes the query scope, the second lists the core capabilities, the third anchors concrete use cases. The purpose is front-loaded and every clause earns its place.

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?

For a 10-parameter read-only query tool with no output schema, the description covers the full action surface — matching modes, filters, collapse, date windows, pagination — and the schema documents each parameter in detail. It does not describe the CDX return format or pagination caps, but those are secondary for a tool whose calling pattern is fully specified.

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?

With 100% schema description coverage, the baseline is 3. The description goes further by linking features to parameters: 'keep one capture per period (collapse)' explains the collapse key, 'page with a resume key' explains resume_key, 'restrict a date window' covers from/to, and 'Filter by status code or MIME type' maps to the filter array. This gives an agent a mental model of how to combine parameters, not just their formats.

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 opens with a precise scope statement — 'Every capture the Wayback Machine holds for a URL, URL prefix, host or whole domain (CDX index)' — establishing a specific query resource and matching scope. The feature list (filter, collapse, date window, resume key) and 'site archaeology' framing clearly differentiate it from the single-snapshot sibling archived_copy.

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 final sentence gives concrete use-case guidance: 'what a site ever had, when a page changed, every URL under a path.' This tells an agent when capture-history queries are appropriate. However, it never explicitly names archived_copy as the alternative for retrieving one archived snapshot, nor does it state when not to use this tool.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.