humane-proxy

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

With no annotations provided, description carries full burden and documents detailed return structure including decision flags (should_escalate, stage_reached) and trajectory tracking purpose of session_id. Could improve by noting privacy implications or side effects of classification.

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?

Uses efficient structured format (Parameters/Returns sections) with zero waste. Every line provides essential information not present in structured fields.

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?

Comprehensive for a classification tool: covers input parameters despite poor schema, documents return dict structure even though output schema exists (adds value), and specifies safety categories. Minor gap in explaining relationship to escalation workflow suggested by sibling tools.

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 has 0% description coverage, but the description fully compensates by documenting both parameters: 'message' is defined as 'user message to classify' and 'session_id' as 'Optional session identifier for trajectory tracking' with default behavior implied.

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 uses specific verb 'Classify' with clear resource 'message' and explicit categories ('self-harm or criminal intent'), clearly distinguishing from siblings get_session_risk (session-level) and list_recent_escalations (retrospective listing).

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?

Usage is implied by the specific classification domain (self-harm/criminal intent), but description lacks explicit when-to-use guidance versus get_session_risk or whether to check every message vs sampling.

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