RegAffairs AI

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

Adds substantial context beyond the annotations: the possible ~1 minute latency, the async fallback returning status "running" plus an answer_id for regaffairs_get_answer_status, that sources are 1,400+ regulator datasets with linked citations, and that each call consumes one question on the user's plan and is saved to history. The only gap is that nothing warns whether an in-flight answer can be reused or re-asked without consuming another credit.

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 and exclusions before the source inventory; every sentence carries routing or behavioral information. The long list of covered regulators and source names is the least essential portion and could be trimmed given regaffairs_list_covered_regulations exists, but nothing is redundant.

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 an open-world, quota-consuming research call with an output schema, the description covers routing, latency, the async status contract, credit cost and persistence. An agent has everything needed to decide to call it and to handle both the immediate-answer and running cases correctly.

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 coverage is 100%, so question, industry and jurisdictions are all already documented in the schema, including the ISO-code example format and the advice to include CAS/EC numbers and market. The description reinforces the question style via examples but adds no parameter-level guidance beyond what the schema provides, so 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 specific verb+resource (answers a regulatory question) with an explicit scope: questions needing reasoning across several sources, jurisdictions or industries, or ones no industry tool covers. It names the sibling family it is not meant to replace (single-fact industry checks) and gives concrete example phrasings, so an agent can route correctly without opening the 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 explicit when-to-use criteria (multi-source, multi-jurisdiction, cross-industry reasoning), an explicit when-not-to-use list enumerating exactly the single-fact queries the check_* siblings handle (list status, classification, exposure limit, ingredient status, MRL, register entry), and excludes legal advice, trial design and submission strategy.

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

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

Annotations already declare readOnlyHint=true, idempotentHint=true and destructiveHint=false, so the safety profile is covered. The description adds genuinely useful behavioral context by disclosing the exact data sources consulted (BPR, Article 95, BPC opinions, HSE/GB, SIMMBAD, ANSES, BAuA, EPA List N, Health Canada), which tells the agent what answers are grounded in.

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-loads the core purpose, then organizes the source list into labeled groups (EU / National / US / Canada), which makes the length scannable. The jurisdiction-by-jurisdiction register enumeration is dense and borders on over-long, but each group carries real routing information.

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 an output schema present, return values needn't be described, and the description fully covers scope, supported jurisdictions, input convention, and the exclusion boundary. For a 6-param, zero-required, multi-jurisdiction lookup tool, 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 83%, so the schema does most of the work (baseline 3). The description adds meaning by framing the PT1–PT22 product-type range, explaining that target_pathogen drives US disinfectant lists, and tying include_suppliers to the Article 95 list, which goes beyond the field-level descriptions.

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?

Opens with a specific verb+resource ('Checks whether a biocidal active substance is approved') and enumerates the exact facets covered (active substance approval, Article 95 list, product authorization, EPA disinfectant lists). It explicitly carves out the sibling regaffairs_check_pesticide_approval_and_mrl, so an agent can distinguish it 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?

States the input convention ('Give an active substance or a product name') and an explicit exclusion with the alternative tool named for plant protection/MRLs. It lacks guidance on choosing among the sibling check_* tools (e.g. check_chemical_substance) or when to set jurisdictions, so it stops short of full when/when-not coverage.

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

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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, so the safety profile is covered. The description adds genuine behavioral context beyond the annotations: the identity-resolution step, the 50-substance cap, and the important semantic rule that absence from a list is reported as 'not listed'. It stops short of describing return format or partial-failure behavior.

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?

Purpose and delivery are front-loaded in the first sentence, and the exclusions close the paragraph cleanly. The middle enumeration of every inventory and list is lengthy and somewhat padded, but it directly informs list selection, so it earns most of its space.

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?

An output schema exists, so return values need not be explained, and annotations carry the safety profile. Combined with full scope, jurisdiction/list coverage, the 50-item limit and explicit sibling boundaries, an agent has everything needed to invoke this correctly.

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 coverage is 100%, so the baseline is 3. The description nevertheless glosses the meaning of `lists` (naming the actual regulatory mechanisms behind the enums) and `identity_only`, and confirms the accepted identifier formats (name, CAS, EC) — adding real meaning over the schema's terse per-field text.

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 states a specific verb+resource ('Checks whether one or more chemical substances... are registered, listed, restricted or banned') and scopes it further ('screens a bill of materials'). It also names the sibling tools it is NOT (GHS hazard classification, workplace exposure limits, industry-specific ingredient checks), so an agent can differentiate it without opening schemas.

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?

Explicit when-not guidance is given: 'Not intended for hazard classification, pictograms or UN numbers', exposure limits, and cosmetic/food/pesticide/biocide status, each with the sibling tool named as the alternative. The 'Accept up to 50 substances' and 'missing from a list = not listed' rules set invocation expectations clearly.

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

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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is covered. The description adds substantive behavioral context: the authoritative sources consulted (Annexes II–VI, Hotlist, ASEAN Directive, etc.), the jurisdictional boundaries, and that the glossary is a specific edition ('last published edition'). It omits freshness/versioning caveats and error behavior, keeping it short of a 5.

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 answer to 'what does this do' in the first clause, then coverage, then exclusions. The regulation enumeration is long, but each item is a genuine coverage claim rather than filler; a mildly tighter grouping could shorten it without loss.

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 an output schema present, return-value documentation is unnecessary, and annotations cover safety. The description supplies scope, source authority and exclusions – everything needed to decide to call it. The residual gap is the unexplained non-enum parameters, which the schema's sparse descriptions do not fully close.

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 50%, so the description carries some burden. It implicitly maps the jurisdictions enum by naming the specific regime for each bloc (EU, UK, US, Health Canada, ASEAN, Japan) and hints at product_type via 'product-type conditions', but it says nothing about the 'function' enum, 'application', or 'for_children_under_3', which are left to the schema alone. Partial compensation warrants a baseline 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?

Opens with a specific verb and resource ('Checks whether a cosmetic ingredient is banned, restricted or permitted') and enumerates the outputs it returns (INCI name, max concentration, product-type conditions, label warnings). The scope sentence pinpoints the exact regulatory domains covered, which cleanly separates it from siblings like check_chemical_substance or check_food_additive_or_ingredient.

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 an explicit boundary: 'Not intended for food, medicines or general chemical restrictions (the food, pharma and chemical tools cover those)' – naming both the exclusion and the alternatives. Combined with the coverage list, an agent can route an ingredient query to this tool or a sibling without guessing.

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

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

Annotations already establish readOnly, idempotent, non-destructive and closed-world behavior, so the safety profile is covered. The description goes further by disclosing exactly which data sources back the answer per jurisdiction and per facet (Orange Book, EMA, shortage lists, MAUDE, NANDO, CDSCO, etc.), which is meaningful coverage/trust context beyond the structured fields.

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 clause, then organized as Medicines / Devices / Not-intended. It is dense and list-heavy, but each listed source maps to a data facet or enum value, so most of the length earns its place. Minor redundancy between the title and opening sentence.

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 an output schema present, return-value explanation is unnecessary, and annotations cover the safety profile. The description supplies source coverage, jurisdictional scope and facet mapping, leaving only edge behavior (no-match results, cross-kind ambiguity) unstated. Adequate-to-strong for a tool of this scope.

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 coverage is only 50%, so the description must compensate, and it does: the enumerated sources map directly onto the `include` enum values (shortage lists, orphan designations, MAUDE adverse events, NANDO notified bodies, FDA device classification), giving the agent a basis for choosing facets. `product` and `jurisdictions` already carry schema descriptions, and the medicine-vs-device framing clarifies `product_kind`, though it doesn't state input formats or behavior for ambiguous products.

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 (checks/retrieves) plus the two distinct resources it spans — a medicine's authorisation/shortage status and a device's classification, registration, approval or notified body. The description lets an agent distinguish it from siblings like regaffairs_check_chemical_substance or regaffairs_check_pesticide_approval_and_mrl without opening any 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?

Provides explicit exclusions and names the alternative: 'Not intended for GMP or ICH guidance (regaffairs_find_pharma_gmp_and_ich_guidance covers that), for MDR or IVDR legal requirements, or for clinical trials.' That is clear routing guidance. It lacks a positive 'use this when' phrasing relative to other checking siblings, so it stops short of a full 5.

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

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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety profile is covered structurally. The description adds genuine behavioral context beyond that: the exact regulatory corpus consulted (1333/2008 Annex II, 2015/2283, 2023/915, 10/2011, 21 CFR 170-189, GRAS notices) and that results are jurisdiction- and category-conditional. It does not discuss latency, rate limits, or answer-completeness caveats.

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?

The purpose sentence is front-loaded and efficient, but the middle is a dense, unbroken enumeration of regulations that reads more like documentation than selection guidance. It is informative yet disproportionately long relative to the decision it supports, and the exclusions would be easier to scan if split out from the coverage dump.

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 complex multi-jurisdiction regulatory lookup with 5 parameters, an output schema, and rich annotations, the description supplies the domain scope, jurisdiction coverage and sibling boundaries an agent needs. It is nearly complete; only the meaning of the 'any' regime default and how multiple jurisdictions combine in the response are left to the schema/output.

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 coverage is 60%, so the description must carry part of the load, and it does: the coverage list maps directly onto the regime enum values (food_additive, colour_additive, novel_food, supplement_or_fortification, food_contact, contaminant, gras), 'EU or US' explains the jurisdictions array, and 'food categories and maximum levels' and 'plastics food contact' imply food_category and contact_material usage. It does not spell out the 'any' default or the accepted input formats beyond the schema's own E-number/INS/FD&C/CAS note.

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 states a precise verb ('Checks whether... is allowed') and enumerates the exact resource classes covered (food additive E-number, colour additive, novel food, vitamin/mineral, food contact substance, contaminant limit) across two jurisdictions. It also names the siblings it is NOT, so an agent can separate it from regaffairs_check_food_label_and_allergens and regaffairs_check_pesticide_approval_and_mrl without opening a 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?

Explicit when-not guidance is present with named alternatives: 'Not intended for labelling, allergen or health claim questions (regaffairs_check_food_label_and_allergens covers those) or pesticide residues (regaffairs_check_pesticide_approval_and_mrl).' Combined with the positive statement of scope (EU or US, allowed and under which conditions), the routing decision is fully determined.

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

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

Annotations already declare readOnlyHint=true, openWorldHint=false, and idempotentHint=true, so the safety profile is fully covered. The description adds useful scope detail (which regulatory regimes it covers) but no behavioral context beyond what annotations provide, such as latency, rate limits, or what happens when a jurisdiction is unsupported.

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?

The description is long but front-loaded: the opening sentence states the three question types, then the middle enumerates regulations, then the final sentence handles exclusions. It is dense but every sentence serves routing or scope. Slightly verbose in the regulation enumeration, which could be trimmed.

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 read-only lookup tool with an output schema present, the description covers scope, jurisdictions, and exclusions adequately. It doesn't explain return format, but the output schema handles that. The regulation enumeration is thorough, though jurisdiction parameter semantics remain thin.

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 50%: topic enum and detail are self-documenting, but food_type and jurisdictions lack full schema descriptions. The description mentions covered regimes (EU, GB, US, AU-NZ, IN) implicitly through the regulation list, but doesn't clarify parameter usage. With 50% coverage, the description adds marginal value only.

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 opens with a precise statement of what questions it answers (mandatory particulars, allergen declaration, claim authorisation), which cleanly distinguishes it from the sibling tools it names explicitly (food_additive_or_ingredient, cosmetic_ingredient). An agent can route correctly without opening any 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?

Explicit negative guidance: 'Not intended for whether an ingredient or additive is allowed (`regaffairs_check_food_additive_or_ingredient` covers that) or for cosmetic labels (`regaffairs_check_cosmetic_ingredient`).' Both exclusions name the alternative sibling tool. This is textbook when/when-not guidance.

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

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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false, so safety is covered. The description adds genuinely useful behavioral context by enumerating the underlying data sources and regulations (Reg 1107/2009, Reg 396/2005, 40 CFR 180, ANVISA, Codex), telling the agent what kind of authority backs each answer. It stops short of freshness/coverage limits for each source.

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 purpose, then the jurisdiction/regulation survey, then exclusions last — a sensible order. It reads as a dense run-on sentence, but each clause maps to a selectable jurisdiction or data type, so little is padding.

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?

An output schema exists, so return-value explanation is unnecessary; the description focuses correctly on scope, coverage and exclusions. For a 4-parameter tool it is nearly complete, with the main gap being the absence of guidance on combining `include` and `jurisdictions` or on when a CAS versus a name should be supplied.

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 50% — only `crop` and `active_substance` are documented, while `include` and `jurisdictions` are not. The description partially compensates by enumerating approval/review/emergency-authorisation/MRL concepts and naming EU, GB, US, Canada, Australia-NZ, Codex and Brazil, which maps to the two undocumented enums, but it never explains them as parameters.

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 precise verb and resource — 'Checks whether a plant protection active substance is approved, banned or under review' and returns the MRL/tolerance for a crop. The scope (pesticide active substances) is unambiguous and clearly separable from the biocide sibling tool it names.

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?

Explicitly excludes adjacent domains: 'Not intended for disinfectants and other biocides (`regaffairs_check_biocide_active_substance` covers those) or for industrial chemicals.' This names both the boundary and the alternative tool, which is exactly the routing information an agent needs.

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

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

Annotations already declare readOnly/idempotent/non-destructive/openWorld=false, so the safety profile is covered. The description adds real scope context — the exact standard lists and labelling regulations consulted — which is not derivable from the annotations. No auth or rate-limit detail, but that is minor here.

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?

Purpose, scope and the sibling exclusion are front-loaded; the long enumerations of lists and labelling rules each earn their place by anchoring schema enum values. The mid-sentence 'Give substances or a topic' is slightly awkwardly placed but not wasteful.

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 an output schema present, return values need not be described. The description covers the substance-list scope, the labelling scope, the input convention and the sibling boundary — sufficient for a 5-parameter, 0-required 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 80%, so baseline is 3. The description exceeds that by mapping enum values to their real-world meaning (zdhc_mrsl = ZDHC MRSL, oeko_tex_100 = Standard 100 class limits, reach_annex_xvii = Annex XVII textile entries) and clarifying the substances-vs-topic interplay, which the enum labels alone do not convey.

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 (checks) and resource (restricted substances and labelling for textiles/clothing/footwear/leather) and enumerates exactly which regimes are covered. It explicitly distinguishes itself from sibling `regaffairs_check_chemical_substance`, so an agent can route 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 an input convention ('Give substances or a topic') and an explicit exclusion with the named alternative ('Not intended for general chemical status outside textiles; regaffairs_check_chemical_substance covers that'). It lacks a positive when-to-use example, but the routing guidance is clear.

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

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

Annotations already declare readOnlyHint/idempotent/openWorldHint=false, so the safety profile is covered. The description adds genuine non-annotation context: it discloses the return shape ('passages with the guideline code, section and version date') and the fact that licence results come from the MHRA GMDP register. It does not note rate limits or result caps, so it falls short of a 5.

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?

Purpose and the 'give a topic/code/site' instruction are front-loaded, and the exclusion clause is placed last where it belongs. The long enumeration of covered corpora is dense but each item aids source routing; it is slightly bloated but not wasteful.

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 4-param, openWorld=false lookup tool with an output schema, the description covers purpose, accepted inputs, exclusions with named alternatives, and (redundantly) the return shape. An agent has everything needed to select and call it correctly.

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 coverage is 75%, so baseline is 3. The description adds meaning beyond the schema by mapping its coverage list onto the 'source' enum values (eu_gmp, us_cgmp, ich, ctd, eu_medicines_law, mhra_licences) and clarifying that site_or_company drives the MHRA lookup, which the bare schema does not spell out.

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?

Names a specific verb ('Finds') and precise resources (GMP/ICH guideline requirements and text, MHRA GMP/GDP licence status), and enumerates the exact corpora covered (EudraLex Vol 4, 21 CFR 210/211, ICH Q/S/E/M, M4 CTD, MHRA GMDP register). An agent can distinguish it from every sibling, especially check_drug_or_device_approval, without opening a 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 explicit input guidance ('Give a topic, a guideline code or a site or company'), explicitly disambiguates from the ask_question sibling ('It does not count as a question'), and states three concrete exclusions (medicine authorisation/shortage, clinical trial conduct/GCP, FDA submission strategy) with the alternative tool named. Nothing is left to inference.

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

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), yet the description still adds real behavioral detail: a bounded wait of up to 40 seconds, two possible outcomes (finished answer with citations, or running again), and that the call does not consume a question. It stops short of saying whether repeated polling is expected or rate-limited, which is the only meaningful 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?

Three short sentences, front-loaded with the verb and resource, then the wait semantics, then the exclusion. No filler and no repetition of annotation facts.

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?

A one-parameter polling tool with an output schema and full annotation coverage needs little more than trigger condition, timing behavior, and what comes back. All three are present, and return-value detail is properly left to the output schema.

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 coverage is 0% and the single property is a bare string, but the description supplies the provenance of answer_id (returned by regaffairs_ask_question alongside status: "running"), which the schema does not say. That is the one thing an agent needs to populate the field correctly.

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 (fetches) and resource (result of a running answer), and explicitly ties it to the sibling regaffairs_ask_question that produces the answer_id. The closing sentence distinguishes it from starting a new question, so an agent can tell it apart from its nearest sibling without opening any 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 the precise trigger condition: use it after regaffairs_ask_question returned status: "running" with an answer_id. It also names the excluded case (not for starting a new question), leaving nothing to inference.

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

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

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is covered; the description adds meaningful context beyond that by naming the underlying sources (ECHA harmonised list, Regulation 1272/2008, C&L Inventory, HSE, Korean GHS, US DOT 49 CFR 172.101) and by warning that mixtures require the user's own calculation. It does not describe coverage limits, staleness of the lists, or failure behavior when a substance is unmatched.

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?

Purpose and scope are front-loaded and no sentence is filler, but the prose is dense and list-heavy, and the UN number/transport concept appears twice (once in the opening clause, once in the transport clause), which slightly inflates the length for a single-parameter-required lookup.

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 five-parameter lookup with an output schema present, the description supplies everything an agent still needs: what sources are queried, what the optional flags add, which jurisdictions are supported, and what the tool deliberately does not cover. Nothing material 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 coverage is already 80%, yet the description adds real semantic value by mapping `jurisdictions` values to concrete data sources (EU CLP Annex VI, GB mandatory list/HSE, Korean GHS) and by tying `include_transport`, `include_notified` and `include_pending` to the specific datasets they pull in. The `substance` identifier formats are left to the schema.

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 ('looks up') and resource (a substance's GHS/CLP hazard classification, label elements and UN number), then enumerates exactly what comes back (Annex VI entry, H-statements, pictograms, signal word, SCLs, M-factors, ATP). It explicitly carves itself out from `regaffairs_check_chemical_substance`, so an agent can distinguish it from siblings without opening the 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 concrete triggers ('for SDS sections 2 and 3 or a CLP label', 'SDS section 14') and explicit when-not conditions: not for mixtures, not for ADR/IMDG/IATA rules, and restrictions go to `regaffairs_check_chemical_substance`. The only gap is that the positive trigger is embedded in the purpose sentence rather than framed as a standalone 'use this when' rule.

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

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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds genuinely useful coverage context by naming the specific national frameworks and databases held (HSE EH40, TRGS 900/GESTIS, INRS VLEP, OSHA Z-1, etc.), which is not derivable from annotations or the country enum alone.

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?

Purpose and exclusions are front-loaded, but the middle is a single sprawling sentence enumerating sixteen source documents, which is longer than selection requires. It is serviceable but not tight; a representative subset plus 'and other national OEL frameworks' would carry the same selection signal.

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?

An output schema exists, so return values need not be explained. The description covers scope, jurisdictional coverage, defaults and exclusions; only minor gaps remain, such as whether limits are returned in normalized units across jurisdictions.

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 country enum, substance identifiers and limit_types defaults are already documented in the schema. The description adds only marginal parameter meaning via the limit acronyms in the title (MAC, STEL, AGW), which loosely maps to limit_types, so the 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 specific verb ('looks up') and resource ('occupational exposure limits for a substance'), enumerates the limit-type acronyms it returns, and names the siblings it is not (search_regulations, get_ghs_hazard_classification). An agent can distinguish it from every sibling without opening a 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 concrete use cases (SDS section 8, COSHH assessment, OSHA compliance check) and explicit exclusions: drinking water/ambient air/food limits route to regaffairs_search_regulations, hazard classification routes to regaffairs_get_ghs_hazard_classification. Both when-to-use and when-not are stated.

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

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

Annotations already declare readOnly/idempotent/non-destructive, so the bar is lower, yet the description adds real context: it returns the *live* catalogue, it validates availability before answering, and it carries no question-quota cost. It does not mention rate limits or pagination behavior.

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-loads purpose in the first clause, then examples, then return shape, then filters, then the quota caveat. Tight overall; the parenthetical example questions are useful but make it longer than strictly necessary.

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 an output schema present, the description needn't detail return values, yet it summarizes them reasonably. For a zero-required-parameter discovery tool the description is nearly sufficient; only the limit parameter handling is left unexplained.

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 names the filter axes ('industry, jurisdiction or a text query'), covering three of the four parameters, but says nothing about the limit parameter or the array format of jurisdictions, leaving genuine gaps.

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 ('Lists what RegAffairs AI covers') and enumerates the exact return fields (dataset id, name, regulator, jurisdiction, industry, contents, last refresh). It is clearly distinguishable from siblings like search_regulations and ask_question because it describes a coverage catalogue rather than a lookup or a question.

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 concrete trigger questions ('do you have the EU Cosmetics Regulation?') and states the operational rule that it 'does not count as a question,' which distinguishes it from regaffairs_ask_question. It does not explicitly name the sibling tool or state when not to use it, so it stops short of a 5.

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

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

Annotations already cover readOnly/idempotent/openWorld safety profile, but the description adds genuinely new behavior: ranked full-text search, passages returned with citations, and per-dataset language search with a concrete non-English example. It doesn't discuss result limits or pagination, but with an output schema present the return-shape burden is lower.

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 purpose and route guidance, and the examples are useful, but the paragraph is dense and runs long. Some sentences are clumsily worded (e.g. 'It does not count as a question') and could be tightened without loss.

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 6-parameter, 50%-coverage tool this is fairly complete: it explains the search model, language behavior, dataset scoping and the document_id workflow, and an output schema covers return values. The main gap is the unexplained industry/jurisdictions/limit parameters.

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 50%. The description adds meaning for query, document_id and the language-to-use-per-dataset rule, but says nothing about industry, jurisdictions, or the limit cap (default 8, max 20). It compensates only partially for the uncovered half.

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 (searches) and resource (wording of regulations, guidelines, register entries) with concrete scope and named datasets. It explicitly distinguishes itself from siblings by saying it differs from a 'written answer' (regaffairs_ask_question) and from 'an industry tool'.

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 clear when-to-use (need the actual text, or an area no industry tool covers like REACH/CLP/EPA/E-PRTR) and when-not ('Not intended for questions an industry tool fits; those tools return structured results'). It also explains the document_id follow-up read pattern. Nothing is left to inference.

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