Agent memory

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

The description and schema reveal genuinely destructive behavior: 'forget' can delete a memory and omitting id with a namespace can wipe it. This contradicts annotations.destructiveHint=false, which tells the agent the tool is non-destructive. Per the rubric, this contradiction forces a score of 1 despite the description adding useful context about credit cost and no vector DB.

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 short sentences front-load the actions and then add the two differentiators that matter operationally: no vector DB to run and wallet credit cost. Every clause earns its place and there is no filler.

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 7-parameter tool, the description plus the fully-covered input schema is mostly sufficient, but the description does not characterize the result of a search/recall payload, and there is no output schema. It also relies entirely on the schema for the important namespace-wipe behavior, so a slightly more complete op-level description would round it out.

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%, with each parameter tagged by operation (store/search/forget), defaults, and enum values. The description adds no parameter-specific detail beyond the action names, so the baseline 3 is appropriate.

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?

Description enumerates three concrete actions (store, search, forget) on a specific resource (agent memories) and adds namespace scoping ('semantic recall on your own namespace'). This distinguishes it from the document-processing siblings in the parserail family, so an agent won't confuse it with parserail_summarize, parserail_extract, or similar tools.

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 clear context: memories are per-agent, recalled semantically, and require no vector DB, so an agent can infer when to use this over other parserail tools. It does not explicitly state when not to use it or name an alternative memory tool, but none appears in the sibling list, so exclusions are less critical.

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