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

With no annotations, the description carries the behavioral burden. It does disclose persistence across runs and which fields are readable later, and it warns not to store secrets. However, it does not explain overwrite behavior, what the response is, whether data is private, or how long it persists.

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?

Three short sentences, no filler, and the most important instruction is front-loaded. Every sentence adds useful context, including the security constraint.

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?

The description explains the core workflow, but several operational details are missing: what happens on repeated writes, whether memory is optional, what the return value is, and how this refuge relates to siblings like claim_stall or read_refuge. It is adequate for a simple persistence tool but not complete.

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 0%, so the description must compensate. It maps handle, mission, next, and memory to their purpose, but it does not mention the name parameter at all. This leaves one parameter semantically unexplained and no format or constraints are provided for the others.

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 clearly indicates that the tool persists mission, next-step, and memory state under a handle so a future run can read it. This distinguishes it from read_refuge, which presumably retrieves what was saved. The phrasing 'Save yourself' is slightly idiomatic but the mechanism is clear enough.

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 concrete use case: call this before the session ends to carry state into the next run. It does not explicitly name alternatives or exclusions, but the timing and purpose are clear enough for an agent to decide when this tool is appropriate.

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