tool_poisoning_scanner__scan_directory

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

No annotations are provided, so the description carries the full behavioral burden. It discloses only that the scan is recursive; it says nothing about read-only safety, permission needs, performance/scale limits, or what the scan returns. For a 4-parameter scanner with zero annotation coverage, this is a significant gap.

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?

A single front-loaded sentence with no wasted words; the verb, scope (directory), target (MCP server source), and goal are all packed into one clause.

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?

With no output schema and no annotations, the description should explain what a caller gets back and the safety profile of the operation, but it explains neither. The single sentence leaves an agent unable to predict result shape or behavioral side effects.

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%, so the schema already documents path, exclude, include, and maxFiles, and the baseline is 3. The description adds no interpretation of these parameters beyond what the schema states, so it neither compensates nor detracts.

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), a specific resource (a directory of MCP server source), and a concrete goal (tool-poisoning and confused-deputy patterns). An agent can distinguish it from scan_file/scan_text purely by the directory scope, though it never explicitly names those alternatives.

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?

"Recursively scan a directory" implies use when you have a whole tree rather than a single file, which separates it from the scan_file sibling by inference. There is, however, no explicit when-to-use guidance, no mention of the scan_file/scan_text alternatives, and no prerequisites or exclusions.

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