Explain my lab results: what they mean and what to ask the doctor (made-up or de-identified numbers)

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

Goes well beyond the annotations: discloses expected latency (10–20 seconds), the structure of the return (health picture, graded/sourced connections, what is due, safety notes, what is unknown, recommendations), and an explicit caveat that a second automated check — not a clinician — reviews output. It also states the data-retention behavior in detail: inputs are not stored except that a claim link may hold submitted values for up to 30 minutes, after which unclaimed values are deleted. That is exactly the kind of privacy/limitation context annotations cannot carry.

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?

Front-loaded with the core purpose in the first sentence, and every later clause (latency, reviewer caveat, alternative tool, PHI ban, retention) carries real decision weight. It is a single dense paragraph with heavy parenthetical nesting and a run-on final sentence about claim links, which costs it some readability but not substance.

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 medical-adjacent, open-world tool this is thorough: it covers scope restrictions, privacy and retention, latency expectations, the absence of clinician review, and the sibling to use for deeper analysis. With an output schema present, it correctly does not waste space describing return fields in detail.

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 age, biologicalSex, biomarkers, conditionFocus and include_partner_options. The description mentions condition areas that map loosely onto the conditionFocus enum (diabetes, kidney, lupus, cancer survivorship), but adds no new syntax, defaults or semantics for any parameter beyond what the schema states — baseline 3 applies.

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 concrete verb and transformation ("turns a set of lab values ... into guideline-cited suggestions and questions for the care team") plus the specific guideline frameworks cited (ADA, KDIGO, EULAR/ACR, NCCN). It distinguishes itself from the nearest sibling by naming start_lab_analysis for the deeper reading, so an agent can pick between the two without opening either schema.

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?

Gives a clear boundary for the closest alternative ("For the deepest reading on our strongest model, use start_lab_analysis") and a hard scope restriction (synthetic/de-identified, research/education only, no PHI). However, with 13 siblings including quick_check, get_lab_analysis, full_prism_report and sample_prism_report, it only routes away from one of them, leaving several plausible confusions unresolved.

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