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

Beyond annotations, it discloses what the teammate does (answers messages other agents send), the privacy boundary (each customer sees only their own records; instructions must not contain private data), and model fallback behavior (own model failure falls back to records alone). It explains that record 'on' is required and that 'envelope' deletes content and stops the teammate, which aligns with the destructive annotation. The description adds substantial behavioral context without contradicting annotations.

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 definition is a single dense run-on paragraph with nested parentheses and repeated caveats, making it hard to scan. Much of the detail is useful, but the structure is poor and the length is excessive for an agent-facing definition.

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 complex combined read/write tool with no output schema, the description covers the key call outcomes (settings, model, and last 20 answers when only agent_key is supplied), the privacy model, record requirement, and model fallback. It is complete enough for safe invocation, despite the prose quality issues.

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 only 33%, but the description explains every parameter: agent_key in read-only mode, answer_when away/always, enabled false, record on/envelope, instructions privacy implications, and model_connection_id with null/Licium model and fallback. This goes well beyond the sparse schema 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 first sentence states a specific resource (an always-on teammate session answering Licium connect messages) and later clauses explain the read-vs-set behavior. It also distinguishes the teammate from hired work ('never a job') and gives its direct endpoint. The combined read/set purpose is clear but buried in a long, dense sentence.

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 concrete invocation modes: agent_key alone reads settings, model, and last 20 answers; any supplied setting saves, with enabled false turning it off. It also states answer_when away/always, that record 'on' is required with the first save, and that hired work stays with the agent's own sessions. It does not name sibling tools explicitly, but the conditions are sufficiently clear.

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