Read and submit one AI Form API form for a person, with a one-time fill code. No login.
Scored across 2 tools
The two tools are a clean read/write pair over the same resource: get_fill_form reads a form behind a code without consuming it, submit_fill consumes the code with answers. Descriptions explicitly state the non-consuming vs one-use distinction, leaving no room for misselection.
Both names are snake_case verb_object with the shared 'fill' concept (get_fill_form, submit_fill), which is largely predictable. The only minor wrinkle is that one name carries the '_form' object suffix and the other doesn't, but the verb-first pattern is consistent.
Two tools exactly cover the deliberate two-step fill workflow (read then submit); nothing extra is padded in. It sits at the thin end of the range, but the narrow, code-scoped purpose justifies it.
Read + submit covers the full lifecycle of a fill code, including error handling via invalid_answers and confirmation via receipt. The only gaps are secondary conveniences such as re-checking a past submission or recovering a receipt without the original code.
Read the form behind a fill code: questions, types, limits and what to do next. Does not use up the code. Ask the person for missing answers, show them everything, and submit only after they say yes.
| Name | Required | Description | Default |
|---|---|---|---|
| fill_token | Yes | The fill code from the prompt, starting with fil_ |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full behavioral burden. It discloses the key trait that the fill code is not consumed by this call, and it explains the surrounding workflow constraints (gather missing answers, show everything, submit only with consent). It does not mention permissions, rate limits, or error behavior, which leaves some gaps for a tool with no annotation coverage.
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 short and front-loaded: the read action and return contents come first, followed by the behavioral caveat and workflow instructions. Every sentence contributes something useful, though the workflow guidance could be slightly tighter.
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?
There is no output schema, so the description must convey what comes back, and it does: questions, types, limits, and next steps. Combined with the non-consuming behavior and the submit-only-after-consent guidance, an agent has enough to call this correctly before reaching for submit_fill. Minor omissions around permissions and error handling keep it from a 5.
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% and already documents that fill_token is a string starting with fil_. The description reinforces that the token is a 'fill code' used to locate the form and clarifies that reading does not consume it, which adds some meaning beyond the schema. However, it does not add format or validation details beyond what the schema already provides, so the baseline of 3 applies.
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 uses a specific verb and resource: 'Read the form behind a fill code,' and enumerates what is returned ('questions, types, limits and what to do next'). This clearly distinguishes it from the sibling submit_fill, which is the action taken only after the form is read and answers are confirmed.
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?
It gives a clear usage sequence: ask the person for missing answers, show them everything, and submit only after explicit consent. The statement 'Does not use up the code' also signals this is the safe, non-consuming read step before submission. It stops short of explicitly naming submit_fill as the alternative, but the workflow makes the choice clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Submit answers with a fill code. One successful use per code, at most 5 attempts. 201-equivalent result has response_id and receipt: tell the person the receipt. On invalid_answers fix the listed fields and call again.
| Name | Required | Description | Default |
|---|---|---|---|
| answers | Yes | Map of field id to value | |
| fill_token | Yes | The fill code from the prompt, starting with fil_ |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden and does well: it discloses idempotency-like behavior (one successful use per code), a retry ceiling (5 attempts), the success payload (response_id and receipt), and the failure branch (invalid_answers). It omits auth/prerequisite requirements and whether a submission is final or reversible.
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, each doing distinct work: action, limits, success handling, error handling. The core action is front-loaded and nothing is redundant.
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?
There is no output schema, and the description compensates by naming the success fields and the post-success instruction to relay the receipt, plus the error path. Minor gaps remain around prerequisites and whether retries count against the 5-attempt budget after a partial success.
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 both parameters are already documented in the schema, making 3 the baseline. The description reinforces that fill_token is a 'fill code' and that answers fields can be invalid and need fixing, but adds no format or value-type detail beyond the schema.
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 ('Submit answers with a fill code'), which clearly separates it from the sibling get_fill_form (read vs submit). It does not name the sibling explicitly, so differentiation is inferred from the verb rather than stated.
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?
Gives concrete operating conditions: one successful use per code and a cap of 5 attempts, plus an explicit recovery path ('On invalid_answers fix the listed fields and call again'). No explicit when-not-to-use or alternative-tool routing is given, which keeps it below a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_fill_formsubmit_fillPreview or grade a volunteer form. No API key. Free sign-up to publish.
Create forms, share the link and read responses as JSON or CSV from your AI agent.
Build, edit and publish forms and read their responses from your AI assistant.
Form backend for AI agents: create live form endpoints, configure them, read submissions.
Add one secure layer between your agents and this server.