Busca semântica na base de conhecimento. Devolve os trechos (chunks) mais relevantes pra usar como contexto, com nome do arquivo de origem e score. Retrieval-only: você sintetiza a resposta a partir dos trechos.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| top_k | No | ||
| account | No |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false. The description adds useful context beyond those annotations by explaining that the tool returns chunks with source filename and score, and that it does not answer directly. This is clear and non-contradictory.
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 concise: two focused sentences in Portuguese. It front-loads the tool's purpose and then states the key behavioral constraint. Every clause contributes meaning without redundant or unnecessary detail.
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?
The description covers the main action, the expected form of results, and the required downstream behavior well enough for simple semantic search with no output schema. It could still be more complete by explaining how top_k/account affect the search and when to use kb_get_document, but the essential task is understandable.
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?
With schema description coverage at 0%, the description needed to explain the meaning/behavior of query, top_k, and account. None of these parameters are described, no defaults or constraints are given, and no semantic value is added beyond the raw parameter names 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?
The description starts with 'Busca semântica na base de conhecimento', naming a specific verb and resource. It also distinguishes itself from sibling tools by describing its output format (source chunks, filename, score) and by explicitly saying it is retrieval-only.
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 description gives clear guidance on when to use this tool: to obtain relevant KB chunks as context and synthesize the answer from them. It states 'Retrieval-only' as an explicit usage constraint, but it does not provide direct when-not-to-use guidance or contrast alternatives such as kb_get_document.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most tools are distinct: kb_* tools handle knowledge base operations, while authenticate/connect/marketplace/toolkit_info serve system-level purposes. However, 'connect' and 'toolkit_info' overlap slightly in reporting connection status, and 'marketplace' is a broad multi-purpose tool that might be confused with 'report_bug' for feedback.
The kb_* tools follow a consistent snake_case verb_noun pattern (kb_add_text, kb_get_document, kb_ingest_url, etc.), but system tools break the pattern: 'authenticate', 'connect', 'marketplace', 'report_bug', 'show_version', 'toolkit_info' use varying conventions (bare verbs, nouns, compound names). The mixed style reduces overall consistency.
With 13 tools, the server covers both the knowledge base domain (9 kb_* tools) and platform management (4 system tools). This is a well-scoped count that doesn't feel bloated, and each tool addresses a distinct operational need.
The kb_* tools cover the core lifecycle: add text, ingest URLs, get/list documents, search, remove. Missing obvious operations like 'update document' or 'create a new empty document' are minor gaps easily worked around. The marketplace tool adds extensibility that mitigates domain gaps, but within the stated purpose the coverage is strong.