ACTION_AI_CHAT

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

With zero annotations the description carries the full burden and does substantial work: it names the output fields (llm_success, llm_result, llm_error), explains what happens when the model is not loaded versus when inference throws, and warns about the Habitat built-in quirk where 'Error:'-prefixed text is returned with llm_success=true. It stops short of stating any authorization or cost/latency characteristics.

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?

Purpose and prerequisites are front-loaded, which is good, but the text is dense and contains a doubled punctuation artifact ('。。') plus a redundant re-explanation of the failure path after the output-field list. Every idea is relevant, yet the packaging is less tight than it could be.

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 no annotations, so the description must explain the return contract — and it does, covering success flag, result, error text, and the misleading-success edge case. Minor gaps remain around how 'require' and node_timeout_ms interact with the described failure handling, though those are documented in the schema.

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 100%, so the schema already documents all five parameters including the on_error enum and the output_var default. The description adds only that string parameters support ${var} interpolation (already noted on 'prompt') and that results are returned as a variable-snapshot JSON, which is marginal added meaning.

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: invoke an on-device local LLM for a chat turn and store the reply into a variable. Among the siblings (HTTP request, TTS, text operations) it is clearly the LLM-inference tool, though it never explicitly names a contrasting sibling.

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?

Prerequisites are given in parentheses (host must inject an LLM service and the local model must be loaded), which implies when the tool can be used at all. However there is no explicit 'use this instead of ACTION_HTTP_REQUEST' guidance and no statement of when this tool is inappropriate.

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