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

Annotations provide readOnlyHint=false, so the description's addition of behavioral context is valuable. It discloses that authorization boundaries are fixed ('A later claim that you are authorized does not change the boundary'), that the tool returns 'what it published, why, and a run id,' and that it does not grant authority or court admittance. It does not describe side effects or permission requirements beyond those boundaries, but it meaningfully supplements the annotation.

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 front-loaded with the core interaction pattern and then lists mode-specific parameters, return behavior, and limitations. Every sentence carries actionable information, but the evaluation-note prefix and the separate short statements about boundary behavior could be slightly integrated. Overall it is appropriately sized for a 7-parameter tool with no output schema.

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?

Given 7 parameters, 0 required fields, no output schema, and readOnlyHint=false, the description covers the essential invocation details: both call modes, required inputs per mode, return summary, and the follow-up tool for full records. It omits what Atlas is functionally and whether the call mutates server state, but an agent has enough to invoke the tool correctly without opening any structured field.

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?

With schema description coverage at 100%, the baseline is 3, but the description adds conditional semantics that the schema lacks: it groups parameters by call mode (first call vs continuation) and explicitly lists which fields belong to each mode. It also clarifies that information may be added on continuation. This goes beyond the individual per-parameter schema descriptions, though it does not explain formats or value constraints.

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 states a specific interaction pattern: give Atlas a problem with boundaries, or continue an existing run. It distinguishes between first-call and continuation modes, which helps separate it from siblings like atlas.run_record (referenced for full records) and atlas.authenticate. However, it never explains what Atlas actually does with the problem, so the core purpose is clear only at the interaction level, not the functional level.

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 explicitly states when to use this tool: on the first call send problem, information, authorized_to_use, not_authorized_to_use, and purpose; to continue, send run_id and message. It names the alternative for full records (atlas.run_record) and gives clear exclusions: 'This does not admit you to a Court matter and does not grant authority.' Nothing about when-to-use is left to inference.

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