GroundRules - US Law Text

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

Annotations already cover safety (readOnlyHint, idempotentHint, non-destructive), so the description earns credit for adding provenance and trust context: it is machine-generated from the build manifest and never hand-typed, and the stale flag reflects a recurring upstream verification. That tells the agent how to weight the data, which no annotation conveys. It doesn't state result size or whether the list can be empty in a fully-hosted environment.

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 sentences, each carrying distinct payload: field inventory, provenance, and absence semantics. The lead phrase 'Per-source coverage' front-loads the purpose efficiently. 'Current-through or vintage' is slightly jargon-heavy but is offset by the immediately following last_verified and stale-flag explanation.

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?

There is no output schema, so the description must carry the return-value burden, and it does: it names every meaningful field and explains the interpretation of the stale flag and of a missing source. For a zero-parameter, read-only listing tool with closed-world scope, nothing an agent needs in order to call it or read it correctly is absent.

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?

The tool takes zero parameters and the schema is trivially complete, so there is nothing to disambiguate. The description correctly spends no words on inputs and instead describes the shape of the returned rows, which is the more useful information here.

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 plus title clearly state this is a per-source inventory list, and it enumerates what each row contains (section counts, current-through/vintage, last_verified, stale flag). It implicitly separates itself from retrieval siblings by framing the result as coverage metadata rather than legal text, though it never names law_search or law_get_section as the tools to use once coverage is confirmed.

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?

Strong negative-result guidance: 'Absence from this list means we do not host that jurisdiction's text, not that no law exists there' tells the agent exactly how to interpret an empty match before falling back to other tools. What's missing is an explicit positive routing statement such as 'call this first to confirm coverage before law_for_location or law_search.'

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, openWorld), and the description adds real behavioral context beyond them: unincorporated areas return county code with no city code, roughly half of municipalities publish no online code, and the response flags that gap. That discloses outcome variability an agent would otherwise misread as failure.

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 verb and the return contents are front-loaded in the first clause, and the second sentence carries a genuinely useful caveat rather than filler. Slightly dense in the parenthetical enumeration of code-source fields, but no wasted sentences.

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?

No output schema exists, yet the description fully enumerates the returned fields (name, publisher, official URL, current-through, whether we host text) and preempts the most likely misreading of a missing city code. Missing only edge behavior such as ambiguous-address handling or what happens with zero inputs (0 required 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 coverage is only 33% (lat/lon undocumented), so the description compensates partially by framing 'address or lat/lon' as alternative input modes. It does not clarify precedence when both are supplied, whether lat and lon must be paired, or the address format beyond 'one-line street address' in 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 ('Resolve') and resource ('full jurisdiction stack: federal, state, county, place') and enumerates what each layer carries. A reader knows exactly what the tool produces, though it never names a sibling to distinguish itself from law_coverage or law_search.

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?

The input forms ('an address or lat/lon') imply the use case of turning a location into its governing codes, but there is no explicit when-to-use, when-not, or pointer to law_coverage/law_search as alternatives. Usage is inferable rather than 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 cover read-only, idempotent, non-destructive, closed-world behavior, so the description's added value is in return-shape disclosure: per-item result/error for batches, provenance fields (source URL, current-through/vintage, retrieval time), parent path and prev/next siblings. It also makes a strong fidelity promise that publisher editorial matter is never carried, which is meaningful behavioral context. It stops short of describing failure behavior for nonexistent ids or truncation of very long sections.

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 retrieval modes, then the batch semantics, then the return contents. The final sentence about editorial matter is slightly editorial in tone but earns its place as a fidelity guarantee; overall there is little waste.

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 no output schema, the description does good work describing what comes back. But for a zero-required-parameter tool with three mutually exclusive-looking inputs, it never states that at least one must be supplied, nor how missing sections or oversized sections are handled.

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% and each parameter is already documented, so the baseline is 3. The description reinforces that references accepts ids or complete citations and is capped at 10, but adds no exclusivity rule or precedence for when id and citation are both supplied.

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: 'Fetch full verbatim text by id or citation' plus batch mode. It is clearly a retrieval-by-identifier tool, distinguishable in spirit from law_search, but it never names the sibling tools so the contrast is only implicit.

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?

'by id or citation, or pass references for a batch of up to 10' lays out the three invocation modes, which implies lookup-after-search usage. However, there is no explicit when-to-use vs law_search/law_for_location guidance, no statement of mutual exclusivity between id, citation and references, and no note on which mode to prefer.

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 safety (readOnly/true, idempotent, non-destructive, open-world), and the description layers substantial extra behavior on top: every snippet carries citation, source URL, retrieval time, and a current-through or vintage label; stale:true marks hits whose source lags upstream; vintage text is a dated snapshot, not current law. It also discloses the meaning_only fallback and that unhosted requested ids surface in a response note.

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 is front-loaded in the first two sentences, followed by scoping, then response semantics. It is dense and the parenthetical jurisdiction example and the stale/vintage sentences both carry weight, but the parenthetical '(e.g. us, us-tx; us-cfr scopes to the CFR alone...)' is partially duplicated by the jurisdictions and sources schema descriptions.

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 8 parameters and no output schema, the description has to explain the return shape, and it does thoroughly: snippet fields, source URL, retrieval time, current-through vs vintage labels, stale flags, meaning_only semantics, and the note listing unhosted ids. Pagination behavior is only implied by limit/offset, which the schema covers, so an agent has what it needs to call and interpret 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?

At 75% schema coverage the schema already documents most parameters, but the description adds real semantics the schema lacks: it explains the difference between us-cfr (CFR alone) and us (US Code plus CFR) for the jurisdiction/source ids, and clarifies that an address is resolved first and then scopes to its stack's hosted sources. Mode and limit/offset get no extra treatment, but they are documented in 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?

Opens with a specific verb+resource ('Search original law') and immediately adds the query surface (topic, question, or citation) plus the fact that exact citations resolve directly. It carves out territory against law_get_section by noting sections over 64KB require get, though law_get_section and law_for_location are never named directly.

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 says to pass address or jurisdictions, not both, and that sources should be used instead of jurisdictions/address. It also gives a clear escape hatch for oversized sections ('sections over 64KB require get'). What's missing is a direct statement of when to call law_for_location or law_coverage instead of using the address/id parameters here.

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