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

Annotations (readOnlyHint=true, idempotentHint=true, openWorldHint=true) already carry the safety profile, so the description rightly focuses on behavior beyond that: the returned column set on every row, the Tanimoto-only-for-similarity caveat, pagination via nextCursor/cursor, and the cross-tool chaining. No contradiction with annotations — 'search', 'find', and 'read the next page' all align with readOnly. Could earn a 5 with explicit remarks on locking/limitations, but the pagination and return-format disclosure already exceed annotation coverage.

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?

Six dense sentences, logically ordered: purpose → modes → requirements → return fields → chaining → pagination. The 'Discovery entry point' framing is front-loaded. It is longer than average, but the tool has 7 parameters and 4 search modes, so the length is earned. Only minor tightening is possible (e.g., max_phase_min is absent from the description but well covered by the schema), so not 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?

Comprehensive for a complex tool: all search modes, input requirements, return columns, chaining targets, and pagination are covered. An output schema exists, so return-value explanation is a bonus rather than a need. Minor gaps — max_phase_min usage in the description and explicit filtering semantics — are fully handled by the parameter schema and output schema, so nothing an agent needs to call it correctly is missing.

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 baseline is 3 and the schema already documents every parameter well (ranges for similarity_threshold, cursor semantics, enum meanings for search_type). The description adds genuine cross-parameter combination logic — which fields are required in which mode, and that only limit may change when re-sending a cursor — which is valuable. But it does not add per-parameter meaning beyond the schema, keeping this at a solid 3.

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 and resource ('Discovery entry point for compounds') plus the two distinct call shapes — name/ID/InChIKey lookup vs. structure search (exact | similarity | substructure). It differentiates from siblings by scope: chembl_search_targets searches targets, and the chain tools chembl_get_bioactivities / chembl_get_drug_info are downstream consumers, while this is the molecule search entry point. A tool like get_assay is clearly distinct. Nothing is ambiguous about what this tool does.

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?

Provides clear conditional context: default search_type=name needs query, structure modes need structure, and 'At least one of query or structure is required.' It also gives an actionable routing instruction — 'Chain molecule_chembl_id into chembl_get_bioactivities or chembl_get_drug_info.' It lacks explicit when-not-to-use or named exclusions against siblings, but the chaining direction and search-selector guidance make usage clear.

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