Extract source table cells without expanding spans. FREE during validation. No key, credits or wallet. Existing source and input limits apply.
| Name | Required | Description | Default |
|---|---|---|---|
| input | Yes | ||
| requestId | No | Optional legacy field; ignored in free mode. | |
| maxPriceMicroUsd | No | Optional legacy field; ignored in free mode. |
| Name | Required | Description | Default |
|---|---|---|---|
| tables | Yes | ||
| spanPolicy | Yes |
Changes observed during successful MCP inspections.
Input schema / properties / maxPriceMicroUsd / descriptionPrevious value: -"Explicit maximum new debit; 0 means recovery only."New value: +"Optional legacy field; ignored in free mode."Input schema / properties / maxPriceMicroUsd / maximumRemoved value: -999999999999Input schema / properties / requestId / descriptionPrevious value: -"Persist the ID returned by agent_utilities_prepare. Reuse after uncertainty."New value: +"Optional legacy field; ignored in free mode."Input schema / properties / requestId / patternRemoved value: -"^\\d{13}_[A-Za-z0-9_-]{16,80}$"Input schema / requiredPrevious value: -[
- "input",
- "requestId",
- "maxPriceMicroUsd"
-]New value: +[
+ "input"
+]Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, openWorldHint=false, and destructiveHint=false, so the safety profile is covered. The description adds real behavioral context: span-expansion behavior, no key/credits required, and that source/input limits still apply. It does not explain output structure, but the presence of an output schema lowers that burden.
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?
Four short sentences, front-loaded with the core behavior. The cost/pricing sentences are relevant to this specific free-mode tool but are somewhat repetitive ('FREE' and 'No key, credits or wallet' say similar things), which is the only minor redundancy.
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?
Covered: read-only nature via annotations, span behavior, cost, and limits. Missing: any statement of what the output contains or what input.html should look like (raw HTML string vs. fragment). With an output schema present and 67% param coverage, the gap is moderate rather than critical.
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 67%, with the two legacy fields documented inline as ignored in free mode. The description adds nothing about the nested input.html object or its 100000 maxLength. Baseline 3 is appropriate since the schema does most of the work, but the description does not compensate for the uncovered portion.
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?
States a specific verb and resource: 'Extract source table cells'. The qualifier 'without expanding spans' meaningfully distinguishes behavior from a generic table extractor. It does not name or contrast with any sibling explicitly, but the resource is specific enough to place it among the html_* family.
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?
No when-to-use guidance and no alternatives named. The only usage-adjacent signal is 'FREE during validation', which addresses cost, not task selection. An agent cannot tell from the description when to prefer this over html_main_text or html_links.
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.