Artigos de lei, súmulas e temas do STF, STJ e TST com o link da fonte oficial. Grátis, sem login.
Scored across 3 tools
The three tools split into identifier-based retrieval (get_precedent, get_provision) and natural-language search (search_legislation), and the descriptions make the precedent vs. statute distinction clear. The only mild overlap is that search_legislation returns both dispositivos and súmulas, so it touches the territory of both getters, but the search-vs-fetch boundary is still understandable.
All three names follow a clean verb_noun snake_case pattern (get_precedent, get_provision, search_legislation), with the verb accurately signaling the operation type. No mixed conventions or ambiguous prefixes.
Three tools is at the low end for a legal-research surface but each one is distinct and justified: search, fetch a provision, fetch a precedent. It is slightly thin — there is no way to list or browse a law's structure — but not mismatched to the narrow 'Pesquisa de Leis' scope.
The core read-only workflow (natural-language search → retrieve exact provision or precedent text with official link and verification status) is covered end to end, which is the full lifecycle for a read-only server. Minor gaps remain, such as retrieving a law's full text/table of contents or looking up a provision by citation without going through search.
Enunciado de súmula do STF, STJ ou TST, ou tema do STJ (repetitivo) ou do STF (repercussão geral), com a questão, a tese e o link da fonte. O mesmo número de tema existe em tribunais diferentes. Cada resultado traz status: verified (conferido no texto compilado do Planalto), unverified (ainda não conferido), under_review (o texto desta base difere do oficial e não é exibido) ou revoked (revogado).
| Name | Required | Description | Default |
|---|---|---|---|
| kind | No | "sumula" (padrão) ou "tema" (repetitivo do STJ ou repercussão geral do STF) | sumula |
| court | Yes | "STF", "STJ" ou "TST" | |
| number | Yes | Número da súmula ou do tema. Ex.: "331", "677" | |
| binding | No | true para súmula vinculante do STF |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
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 genuinely non-structural behavior: each result carries a status of verified, unverified, under_review (text diverges from the official source and is not shown) or revoked, which tells the agent how much to trust and display the returned text. It does not discuss caching, freshness timing, or rate limits.
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 identification of the resource, followed by the cross-court caveat and then the status semantics; every sentence carries information. No filler, though the status enumeration could be slightly tighter.
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 four-parameter lookup with full schema descriptions and an output schema, the description covers what is needed: what the record contains and how to interpret the status flag. Nothing essential is missing, though a one-line pointer to search_legislation for discovery-by-text would have closed the remaining gap.
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 schema baseline is 3, but the description adds meaning beyond it by explaining that the same tema number exists across different courts, making the court parameter a disambiguator rather than a filter. The kind/court pairing (tema only for STJ repetitivo or STF repercussão geral) is reinforced by the resource sentence, though the binding flag is left entirely 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 resource precisely: an enunciado de súmula from STF, STJ or TST, or a tema (repetitivo of STJ / repercussão geral of STF), along with the questão, tese and source link. A reader immediately knows what is returned. It does not explicitly contrast itself with get_provision or search_legislation, but the resource is distinct enough that confusion is unlikely.
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?
Usage is only implied by the resource description; there is no explicit statement of when to call this versus search_legislation or get_provision. The one guidance-like sentence, 'O mesmo número de tema existe em tribunais diferentes', usefully warns that the number alone is not unique and that court selects the correct record.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Texto de um artigo, parágrafo, inciso ou alínea de lei federal brasileira, com o link oficial. Cada resultado traz status: verified (conferido no texto compilado do Planalto), unverified (ainda não conferido), under_review (o texto desta base difere do oficial e não é exibido) ou revoked (revogado).
| Name | Required | Description | Default |
|---|---|---|---|
| law | Yes | Lei: apelido ou URN. Ex.: "CLT", "Código Civil", "CF", "CDC" | |
| item | No | Inciso em romano: "XXIX" | |
| block | No | Só para o ADCT da Constituição: "ADCT" | |
| article | Yes | Número do artigo. Ex.: "11", "7º", "1.022", "11-A" | |
| subitem | No | Alínea: "a" | |
| paragraph | No | Parágrafo: "1", "2º" ou "único" |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover read-only, idempotent, non-destructive, and closed-world behavior, so the bar is lower, but the description adds genuinely useful disclosure: the `status` values and, critically, that under_review text is deliberately not displayed and that revoked content may come back without usable text. It stops short of saying what happens when a provision simply does not exist.
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?
Two tight sentences with the core purpose front-loaded and no filler. The status enumeration is long but each value carries real decision weight for the agent, so almost nothing is wasted.
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?
Because an output schema exists, the description need not explain return shape, and what it does explain (status semantics, official link) covers the highest-risk ambiguity around missing or divergent text. The remaining gap is routing guidance against the two sibling tools, which matters for a six-parameter lookup 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 100% and every parameter carries its own example, so the schema is doing the heavy lifting. The description adds the legal-hierarchy framing (artigo, parágrafo, inciso, alínea) that conceptually maps to article/paragraph/item/subitem, but it never mentions the `law` alias/URN form or the ADCT `block` parameter.
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 resource precisely: the text of an article, paragraph, item, or subitem of Brazilian federal law, plus an official link. It is clear what is returned, but it never contrasts itself with its siblings get_precedent and search_legislation, so an agent must infer the boundary itself.
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?
Use is only implied: the tool fetches one known provision rather than searching. There is no explicit statement such as 'use this when you already know the law and article; use search_legislation to find them', nor any mention of when get_precedent would be the right choice.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Busca dispositivos de lei e súmulas por uma pergunta em linguagem natural e devolve os mais próximos, com o link oficial. Cada resultado traz status: verified (conferido no texto compilado do Planalto), unverified (ainda não conferido), under_review (o texto desta base difere do oficial e não é exibido) ou revoked (revogado).
| Name | Required | Description | Default |
|---|---|---|---|
| law | No | Restringe a uma lei: "CDC", "CLT", "CF"… | |
| query | Yes | Pergunta. Ex.: "prazo para reclamar de vício aparente" | |
| top_k | No | Quantos artigos devolver (padrão 5) |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint=false, so safety is covered. The description adds genuine context beyond that: every result carries a `status` whose four values are defined, including the important caveat that under_review text diverges from the official Planalto compilation and is not displayed. It does not discuss result ordering quality, empty results, or rate limits, which keeps 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?
Two sentences, both front-loaded: the first covers what is searched and what is returned, the second covers result interpretation. The status enumeration is long but each value carries decision-relevant meaning, so nothing is wasted.
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 need not enumerate return fields, and it still adds the one interpretation-critical nuance (status semantics). What is missing is modest but real: guidance on narrowing by `law`, behavior when no close matches exist, and any hint about how proximity is ranked.
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% and all three parameters (law, query, top_k) are documented in the schema itself, so the baseline of 3 applies. The description reinforces that `query` is a natural-language question rather than keywords, but adds nothing about the `law` restriction or `top_k` behavior.
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 concrete verb and resource ('Busca dispositivos de lei e súmulas') and the retrieval mode ('por uma pergunta em linguagem natural'), returning the closest matches with official links. It does not explicitly distinguish itself from the siblings get_precedent and get_provision, even though it overlaps them by touching súmulas and provisions, so an agent must infer that this is semantic discovery rather than direct retrieval.
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?
Usage is implied by the phrase 'pergunta em linguagem natural', which suggests this is the tool for topical/descriptive queries, but the description never says when to prefer it over get_precedent or get_provision, nor does it state exclusions or prerequisites. The agent must derive the routing rule from the sibling names alone.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
get_precedentget_provisionsearch_legislationJurisprudência de 74 tribunais e órgãos, com link oficial: busca, súmulas, temas e verificador.
Acervo jurídico brasileiro: busca, Markdown, grafo de normas e precedentes, citação verificável.
CNPJ, empresas em lista, publicações do DJEN, prazos, editais de licitação e advogados do Brasil.
Pesquisa de legislação e jurisprudência brasileiras. Exclusivo para assinantes MinutaIA.
Add one secure layer between your agents and this server.
