Skip to main content
Glama

ask

Sends a message to a web chat site (or several at once) and returns the first answer, so agents can get free AI responses without an API key. Supports threads, file uploads, and new chats.

Instructions

One message to a web chat. site: glm | qwen | kimi | deepseek | gemini, or several comma-separated ("deepseek,qwen,glm"): the first that answers wins, sites out of quota / needing a human are skipped. message_file: path whose text is sent (after message if both) - cheaper than writing a long prompt inline. thread: name of a conversation to continue. files: absolute paths uploaded into the chat. new_chat: start over. model: a model name from the site's menu instead of the strongest one. max_chars: longer answers are saved whole to answer_file and only the head is returned (0 = everything). Returns {answer (markdown), site, model, url, seconds, new_chat, skipped, answer_file?, answer_chars?}.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "site": "<site>"
    },
    "name": "ask"
  }
}

Input Schema

NameRequiredDescriptionDefault
siteYes
filesNo
modelNo
threadNo
messageNo
timeoutNo
new_chatNo
max_charsNo
message_fileNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A3.9/5.0
Behavior4/5

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

No annotations exist, so the description carries the full burden, and it does disclose real behavior: the first answering site wins, sites out of quota or needing a human are skipped, long answers are saved whole to answer_file and only the head returned, and message_file is cheaper than an inline prompt. Gaps remain: `timeout` is never explained and there is no mention of auth or failure modes beyond skipping.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Every clause adds information and the most decision-relevant fact, which sites to pass and how racing resolves, is front-loaded. The telegraphic register is dense but contains no filler sentences.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The return shape is spelled out field by field (answer, site, model, url, seconds, skipped, answer_file), which matters since no output schema exists, and the multi-site fallback behavior is explained. The unexplained `timeout` parameter and absent permission/auth notes are the only meaningful omissions for a 9-parameter tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 0% across 9 parameters, so the description must compensate, and it does for 7 of them: site semantics, files as absolute paths, thread continuation, new_chat reset, model as a per-site menu choice, and max_chars truncation with 0 meaning everything. Only `timeout` is left undefined and `message` is only implicitly covered via its ordering versus message_file.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a concrete verb+resource ('One message to a web chat') and immediately enumerates the accepted site values, so an agent can tell this is a send-a-prompt-to-a-web-LLM-UI tool. It stops short of naming when to pick it over siblings such as council or thread_history, so the differentiation is only partial.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It implicitly defines usage: continue a conversation via `thread`, start over with `new_chat`, or pass several comma-separated sites to race them. However there is no explicit when-not-to-use guidance and no reference to alternatives like list_sites or thread_history for discovery, leaving routing to inference.

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