memory_poisoning_scanner__scan_text

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

With no annotations, the description carries the full load. It makes clear this is a static content scan and defines the risk classes, but does not state whether code is executed, what findings/severity output looks like, or whether memory references require particular pattern fidelity. Useful risk definition, though behavioral details remain thin.

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?

Single front-loaded sentence beginning with the action and risk name, then elaborating the two patterns. Dense but every clause adds detection specificity; slightly long and unbroken by structure prevents a 5.

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 two-parameter scanner with no output schema, the description defines the risk surfaces thoroughly enough to know what it will catch. The main missing element is a hint about findings/output shape, but the risk definition compensates well.

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 100% and both parameters (content, optional label) are documented in the schema itself, so the description correctly adds no parameter syntax. Baseline 3 applies since the schema already does the work and the description adds nothing param-specific.

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 (scan) and resource (agent code) and then precisely enumerates the two memory-poisoning patterns it detects: unsanitized untrusted input written to persistent cross-session memory, and retrieved memory injected into a trusted system/prompt context. This differentiates it cleanly from sibling scanners like prompt_injection_scanner__scan_text and tool_poisoning_scanner__scan_text.

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?

The detection scope makes it clear this tool is for code involving archival/core memory or long-term vector knowledge bases feeding prompts, which implicitly routes the agent away from the prompt-injection and tool-poisoning siblings. There is no explicit 'use X instead when Y' statement or note on when not to use it, so it falls short of a 5.

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