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

Annotations already state readOnlyHint=false, destructiveHint=false, idempotentHint=false, so the description is not contradicting anything, but it also adds nothing beyond them. The two most consequential behaviors — what the non-idempotent overwrite flag does to existing collection contents, and what the confirm parameter gates (a safety confirmation for a write) — are never mentioned, and the description does not mention that ingestion can be scoped to the administrator-configured root via include_configured_directory.

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?

It is short, but the second sentence ("Manage collection management operations") earns nothing — it restates the parenthetical and actively misdirects toward collection management. Briefness here is under-specification rather than economy.

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 7-parameter non-idempotent write tool with an output schema, the description should at minimum explain the ingestion sources (paths vs inline contents vs configured directory) and the overwrite/confirm safety semantics. Since an output schema exists it needn't cover returns, but the mutation semantics gap is significant.

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 schema already documents all 7 parameters, including the useful note that document_paths are relative paths beneath the configured document root. The description contributes zero additional parameter meaning, which lands at the baseline for high coverage.

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?

"Add documents" names a verb and resource, so the core action is guessable, but the follow-on "Manage collection management operations" muddies it — it describes collection management rather than document ingestion and does not say what kind of documents or where they land. No sibling distinction is offered, though none of the listed siblings appear to ingest documents.

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?

There is no guidance on when to use this versus other approaches, no mention of prerequisites (does the collection need to exist first, per vector_create_collection?), and no explanation of when to choose document_paths vs document_contents or when to set overwrite/confirm. The agent is left to infer everything from parameter names.

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=false, destructiveHint=false, idempotentHint=false, and openWorldHint=false, so the safety profile is largely covered by structured fields. The description adds nothing beyond those annotations: it does not explain overwrite behavior, confirmation requirements, document ingestion side effects, or what happens to existing collections.

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 short and front-loads 'Create collection', but the second sentence is redundant with the already-stated category and does not earn its place. It is concise but structurally 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 collection-creation tool with seven optional parameters, document-path ingestion options, an overwrite flag, and a confirmation parameter, the description is far too thin. Even with an output schema and annotations covering safety, an agent needs more behavioral context about what creation entails and how the optional parameters interact.

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 schema already documents all seven parameters. The description adds no parameter semantics of its own; given the high schema coverage, a baseline 3 is appropriate.

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 first clause names a specific verb and resource ('Create collection'), so the core action is identifiable. However, the parenthetical just repeats a category name and 'Manage collection management operations' is a tautological filler that adds no scope, constraints, or differentiation from siblings like vector_delete_collection or vector_add_documents.

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 no guidance on when to use this tool instead of siblings, no prerequisites, and no mention of overwrite/confirm semantics. An agent only knows it is the create variant from its name, not from any stated usage context.

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?

The description is consistent with the annotations (destructiveHint=true, idempotentHint=true, readOnlyHint=false) but only because it simply says 'Delete collection.' It adds no behavioral detail beyond the annotations—nothing about what data is removed, whether deletion is reversible, or how the `confirm`, `overwrite`, and document-related parameters affect 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?

It is short and front-loads the deletion action, but the second sentence ('Manage collection management operations') is filler that does not earn its place. The first sentence mostly repeats the tool name and title.

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 destructive mutation tool with seven parameters, the description is incomplete: it omits confirmation behavior, deletion scope, and the role of unexpected document-related parameters. Annotations cover the basic safety profile and an output schema exists, but operational context is still 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 schema is the primary source for parameter documentation and the baseline is 3. The description itself adds no meaning about parameters; notably it does not explain why document ingestion parameters appear on a delete tool.

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 restates the tool name ('Delete collection') and adds only a vague domain label plus the tautological 'Manage collection management operations.' It does not distinguish this from siblings such as vector_create_collection beyond the obvious verb, and a reader gains no addition over the title.

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?

There is no guidance on when to use this tool, when not to, or what alternatives exist. For a destructive collection operation, it also gives no prerequisites or confirmation guidance, leaving all routing and safety reasoning to the agent.

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 declare readOnlyHint=false and idempotentHint=false, which signal a non-read, non-idempotent operation that a search tool would normally not be. The description's "Manage search operations" neither resolves nor explains this, and says nothing about what state might change, latency, or whether the index is mutated. With annotations this ambiguous, silence is a real 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?

The definition is short, so it is not bloated, but the second sentence ("Manage search operations") earns nothing and the first is front-loaded with an ambiguous cross-reference. Brevity here reflects under-specification rather than discipline.

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 7-parameter hybrid retrieval tool with weight parameters and RRF fusion settings, the description conveys zero operational information. An output schema exists so return values need not be covered, but nothing about how to combine lexical_weight/semantic_weight, what rrf_k controls, or what query field to populate is present.

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?

Seven parameters exist (rrf_k, lexical_weight, semantic_weight, number_results, collection_name, question, db_type) and the description mentions none of them. Although schema coverage is nominally 100%, the schema descriptions are pure placeholders ("rrf k", "lexical weight"), so no meaning is conveyed by either source – the description should have compensated and did not.

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?

"Lexical search (vector_search)" gestures at a verb and resource, but the parenthetical points at a sibling tool rather than this one, and "Manage search operations" is filler that restates nothing concrete. An agent cannot tell from this whether the tool retrieves documents, configures a search stack, or modifies an index.

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?

No condition is given for choosing this over vector_semantic_search or vector_search, even though the sibling list makes that choice the central decision. Mentioning "vector_search" in parentheses is ambiguous – it could be read as a pointer to a related tool or as the underlying implementation, adding no routable 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=true, idempotentHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is fully covered structurally. The description adds nothing beyond that—it does not say what a collection listing contains, whether it is scoped or paginated, or why a read-only listing accepts overwrite/confirm/document_paths parameters; the "Manage collection management operations" phrasing even suggests mutation, which sits awkwardly with readOnlyHint=true.

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?

Only two sentences, but the second ("Manage collection management operations") is pure filler that restates the title and adds ambiguity rather than information. Brevity here comes from under-specification, not efficiency.

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 described, and annotations carry the safety profile. However, the description is inadequate for a 7-parameter tool with confusing ingest-shaped parameters on a read-only listing operation—an agent has no way to know which parameters are meaningful or how the results are scoped.

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?

Nominal schema coverage is 100%, but the schema descriptions are vacuous restatements of the field names ("confirm", "db type", "overwrite"), so the coverage metric overstates real semantics. The description contributes zero parameter guidance and does not explain why a list operation exposes ingest-oriented fields like document_paths, document_contents, overwrite and include_configured_directory.

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 first clause names a specific verb and resource ("List collections"), which is enough to identify the operation. But the parenthetical "(vector_collection_management)" and the follow-up "Manage collection management operations" blur the purpose into a generic capability-group label, and nothing distinguishes this from sibling tools like vector_semantic_search or vector_lexical_search that also touch collections.

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?

There is no statement of when to call this tool, no preconditions, and no mention of alternatives among the six siblings listed. An agent must infer from the name alone that this is the discovery/introspection call before searching.

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 safety is covered elsewhere. The description adds no behavioral context such as how hybrid ranking works, what the weights control, or what the output contains.

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 short, but both sentences are essentially filler: the first repeats the name/title, and the second says only 'Manage search operations.' This is under-specification rather than useful conciseness.

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 seven optional parameters, an output schema, and sibling search tools, the definition should clarify hybrid search behavior and selection criteria. The description instead provides almost no context for using the tool 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 description coverage is 100%, so by the stated baseline the schema is treated as documenting the parameters even though descriptions like 'rrf k' and 'semantic weight' are minimal. The description itself adds no parameter meaning beyond 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 says only 'Search (vector_search). Manage search operations.' This restates the tool name and title without specifying what kind of search it performs or how it differs from the sibling tools vector_semantic_search and vector_lexical_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?

There is no guidance on when to use this tool versus vector_semantic_search, vector_lexical_search, or any of the collection tools. The agent is left to infer context entirely from the name and schema.

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?

The tool name and description identify a search operation, which is conventionally read-only, but the annotations explicitly declare readOnlyHint=false. The description does not explain this discrepancy, nor does it describe any state changes, permissions, rate limits, or side effects that would justify a non-read-only search operation.

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 short, but the second sentence 'Manage search operations' is vague filler that does not add actionable information. It is concise but not front-loaded with the distinctions an agent needs.

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 an output schema and annotations, and the input schema has full coverage, but the description still omits critical context for a semantic search tool with seven optional parameters. It does not distinguish semantic search from lexical search or vector_search, and it leaves the non-read-only behavior unexplained despite the annotation.

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 schema technically documents all seven parameters. However, those descriptions are tautological ('rrf k', 'db type', 'question'), and the tool description adds no meaning about how rrf_k, lexical_weight, semantic_weight, or number_results should be used. Baseline 3 applies because the schema coverage is complete.

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 names a specific operation, 'semantic search,' but the parenthetical '(vector_search)' is ambiguous and may collide with the sibling vector_search. 'Manage search operations' is vague and does not clearly distinguish this tool from vector_search or vector_lexical_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?

There is no guidance on when to use this semantic search versus the sibling vector_search or vector_lexical_search tools. The description also gives no prerequisites, weighting guidance, or expected use context.

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