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

With no annotations at all, the description carries the full burden and does well: it discloses the deterministic nature of the screening, its limitation as heuristic rather than calibrated or guaranteed, the 64 KiB size cap, and the absence of model calls or automatic follow-on purchase. This gives an agent a realistic picture of side effects and reliability before calling.

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?

Three short sentences, each earning its place: purpose, limits, caveats, and side effects are packed efficiently with the core purpose front-loaded. The only blemish is the ambiguous '$1' prefix, which muddies the opener for no evident benefit.

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?

The output schema exists, so the return format need not be spelled out, and the description adequately covers limits, caveats, and side effects. The notable gaps are that the required agent_id parameter is never explained and the '$1' phrasing leaves unresolved whether a cost implication exists — both matter for an agent deciding to invoke the tool.

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 0%, so the description must compensate. It makes a meaningful contribution for 'texts' by constraining the batch to 1–20 samples and 64 KiB total, which the bare schema does not state. However, 'agent_id' is left completely unexplained even though it is required, leaving the agent to guess its role.

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?

The description conveys that the tool screens text samples for deterministic injection markers, naming the resource (text samples) and the analytic purpose. However, the opening '$1 batch' is confusing (reads like a price or an index), and the description never distinguishes screen_injection from sibling tools like scan_source or security_preflight, so an agent must infer what makes this screening-specific tool different.

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?

There is no explicit guidance on when to call this tool versus scan_source, security_preflight, or get_security_receipt. The caveat that results are 'heuristic indicators, not calibrated probabilities' implies a rough first-pass screening rather than a final safety decision, and 'no automatic follow-on purchase' hints at low cost, but these are inferences, not instructions.

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