memory_save

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

With no annotations, the description must carry the behavioral burden, and it does substantial work: it forbids persisting unverified guesses as 'verified' and bans sensitive credential/PII storage. It is silent on write semantics such as deduplication, conflict handling with existing memories, or whether the saved record is returned with an identifier.

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?

Two sentences, front-loaded with the positive instruction followed by the prohibitions, with zero filler. Every clause either defines what qualifies as durable or prevents a specific misuse.

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 5-parameter, annotation-free mutation tool with no output schema, the description covers what to store, how confidence should be assigned, and what must never be stored. The remaining gap is what happens on save (duplicate handling, whether an existing fact is overwritten), which an agent would need to invoke it confidently.

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 80%, so the baseline is 3, but the description adds genuine meaning: the listing of fact types maps onto the category enum, and 'never store your guesses as verified' explains the practical use of the confidence enum beyond the schema's own gloss. It adds nothing about tags or source, which the schema already covers.

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+resource ('Save a durable fact') and enumerates the fact types (identity/preferences, project details, decisions, lessons), which clearly separates it from retrieval siblings like memory_search and memory_get. It does not explicitly distinguish it from memory_update, which is the closest ambiguous alternative.

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?

Gives concrete inclusion criteria and explicit exclusions ('Never store passwords, API keys, tokens, card/account numbers, or ID numbers'), which is real when/when-not guidance. It omits workflow guidance such as searching for an existing memory before saving or using memory_update to revise one.

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