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

Beyond the readOnlyHint and idempotentHint annotations, the description discloses concrete behavioral constraints: only validated SELECT, read-only WITH, EXPLAIN SELECT, and metadata tables are allowed; writes, arbitrary functions, multi-statements, comments, and execution modifiers are rejected. It also states that credentials remain on the server and that rows and execution time are bounded.

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 compact and information-dense, with no filler. It front-loads the core purpose, then provides prerequisite, allowed/rejected behavior, security, and bounds in a logical order, and every sentence adds value.

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 database query tool with rich annotations, the description covers prerequisite discovery, accepted SQL constructs, rejected inputs, security posture, and bounded execution. This is sufficient for an agent to invoke it correctly without an 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?

The description gives meaningful semantics for connectionId by saying to obtain it from inspect_project, and for query by describing supported and rejected SQL. However, maxRows is not explicitly described, and with 0% schema description coverage the description carries the burden for all parameters; the phrase 'Rows and execution time are bounded' only partially compensates.

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 and resource: 'Read-only query of a discovered PostgreSQL, MySQL or SQLite database.' It clearly distinguishes this from sibling tools like http_request, git_status, and inspect_project by scoping it to database queries.

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?

It explicitly instructs the agent to call inspect_project first for a connectionId, which is a clear usage prerequisite. It also describes what SQL forms are allowed and rejected, but it does not explicitly name alternative tools for cases where db_query is not appropriate.

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?

Whereas annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, the description adds meaningful behavioral guarantees: 'Does not run hooks, modify Git state, return credentials, or expose diff contents'. The worktree limitation also discloses a real edge-case constraint. This goes well beyond the structured annotations.

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 dense sentences cover purpose, usage, and behavioral exclusions with no filler. The most important scoping information is front-loaded, and every clause adds operational value. This is an ideal size for a tool description.

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 thoroughly covers safety, scope, and limitations, but it omits the meaning of the optional 'path' parameter and does not describe the output structure. Since there is no output schema, the agent must infer what the tool returns beyond the listed path categories. Overall it is close to complete but leaves a notable ambiguity in invocation.

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?

There is a single optional 'path' parameter with 0% schema description coverage, yet the description never explains what 'path' means or how it affects behavior. The parameter could refer to a repository path or a file/directory filter, which is ambiguous and potentially material to correct invocation. With low schema coverage, the description was required to compensate 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?

The description opens with 'Read-only Git branch, modified, added, deleted, untracked and staged paths', naming a clear verb, resource, and exact scope. It also positions the tool among siblings by tying it to local Git state and runtime failures, making it easy to distinguish from http_request, db_query, and read_logs.

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?

It explicitly says to 'Use to relate local changes to runtime failures', giving a concrete intended use case. It also provides an exclusion: 'Linked worktrees are currently unsupported'. It does not explicitly name when not to use or name alternative tools, but the guidance is still clear enough for selection.

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?

Beyond the annotations (readOnlyHint=false, destructiveHint=true), the description discloses concrete behaviors: supported methods, no redirects, no credential forwarding, possible side effects even for GET, startup opt-in for other methods, and untrusted response content. This adds substantial operational context and does not contradict the annotations.

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 compact and front-loaded with the core purpose. Each of its five sentences adds a distinct behavioral or scoping fact, with no filler or repetition.

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 mutating network tool with no output schema, it covers scope, enabled methods, side-effect risk, auth behavior, redirects, and untrusted output. The only notable gap is that the schema enum includes POST, PUT, PATCH, and DELETE while the description says enabled methods are GET, HEAD, and OPTIONS; the startup opt-in note mitigates this, but the relationship is not fully explicit.

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 clarifies the meaning of method, body, and headers, and implies that url is the target endpoint. However, it does not explain timeoutMs semantics or the maximum-length constraints, though those are partly visible 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 opens with a specific verb ('Tests') and a precise resource ('explicitly requested HTTP endpoints within the user-configured network scope'). It clearly identifies this as a network-request tool, which is distinct from the sibling tools that inspect git status, databases, processes, ports, and logs.

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?

It states when the tool is appropriate (explicitly requested endpoints in scope), which methods are enabled, and that other methods require startup opt-in. It also gives exclusions: no redirect following and no discovered credentials. It does not explicitly name alternative tools, but none of the siblings are network-request tools, so the guidance is sufficient.

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. The description adds valuable context: it clarifies that detected services are not proof of running processes and emphasizes that files are never executed, which is important behavioral information beyond the annotations. This prevents the agent from over-trusting the output. No contradictions with annotations.

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, with three sentences covering purpose, usage, and key caveats. It front-loads the core purpose and adds important warnings at the end. No filler. It could be slightly more structured, but it is efficient and clear.

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 single-parameter, read-only tool with rich annotations, the description is quite complete. It covers what the tool does, what it might return (languages, scripts, ports, masked connection IDs), and important limitations (expectations vs. running processes). It does not detail output structure or error cases, but given the lack of output schema and simplicity, this is sufficient.

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 schema has only one parameter 'path' with 0% schema description coverage, meaning the schema itself provides no description. The tool description does not explicitly explain what 'path' means, but the tool name and description imply it refers to a project directory. Since there is no explicit parameter breakdown, the description partially compensates by giving context (project discovery) but could be improved by stating 'path' should be a directory path. Given low schema coverage, the description carries some burden but does not fully specify semantics, so a 4 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 description clearly states that this is a read-only project discovery tool and enumerates what it identifies: languages, package scripts, likely service ports, and masked database connection IDs. It uses a specific verb 'inspects' (implied) and resource 'project', and distinguishes itself from siblings like git_status and list_processes by its scope (project discovery) and read-only nature.

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 explicitly says 'Use first to understand configuration', which gives clear guidance on when to use this tool (as an initial discovery step). It does not name alternatives, but given the tool's unique role as a discovery tool among siblings (like http_request, db_query, list_processes), the guidance is clear enough for an agent to know when to call it. The admonition that detected services are expectations, not proof, also helps with usage expectations.

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, and non-destructive behavior. The description adds meaningful context beyond that: ownership visibility depends on OS permissions, and no processes are modified. This informs the agent about environment-dependent output without contradicting the annotations.

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 two compact sentences. The core purpose and read-only nature are front-loaded, and the use case and permission caveat follow without unnecessary detail. Every sentence adds distinct value.

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 simple read-only listing tool with no output schema, the description covers what is listed, when to use it, and a key platform-dependent limitation. The only notable gap is the undocumented port filtering parameter, which an agent would need to know to target a specific port.

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%, and the description does not mention the optional 'port' parameter or explain that it likely filters results to a specific port. The schema only provides type and range constraints, leaving the parameter's purpose and behavior undocumented. A single optional parameter is simple, but the description still fails to clarify how it affects the returned list.

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 clearly states the action ('list') and the resources ('listening TCP and bound UDP ports' and 'owning processes'). It also specifies scope conditions like 'when visible,' making the tool's purpose unambiguous and distinct from a generic process listing.

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 explicitly says 'Use to diagnose an occupied port or a service that is not listening,' which gives a concrete trigger for when to call the tool. It stops short of naming when not to use it or contrasting it with the sibling list_processes, so no exclusions are provided, but context 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 mark it read-only and non-destructive, and the description aligns with those. It adds meaningful context beyond annotations by disclosing that command arguments are omitted to protect credentials, which is a non-obvious behavioral detail.

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 tightly written sentences front-load the core behavior and then add usage guidance and limitations. Every sentence earns its place with no redundancy.

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 output fields, read-only nature, and credential protection, which is good for a simple tool. However, the search parameter's behavior is left entirely unexplained, and there is no output schema to fill that 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?

The only parameter, 'search', has 0% schema description coverage and is not mentioned in the description. The agent cannot determine whether search filters by process name, PID, or something else, and the description does nothing to compensate.

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 the verb 'list', the resource 'local processes', and the returned fields (PID, executable name, CPU, memory). This clearly differentiates it from siblings like list_ports or read_logs.

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?

It explicitly says 'Use to check whether an application is running,' giving a clear when-to-use signal. It also states what it cannot do ('Cannot kill, restart, or modify processes'), but it does not name a specific alternative tool for those actions.

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?

Beyond the readOnlyHint and idempotentHint annotations, the description adds genuinely useful behavioral context: the search scope limit (last 1 MiB per file) and the security warning that returned log text is untrusted data. This materially shapes how an agent should invoke and interpret the tool.

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?

Four short sentences, each carrying distinct information: operation and purpose, use case, limitations, and data trust. It is front-loaded with the core behavior and contains no filler.

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?

Given no output schema and sparse parameter descriptions, the description covers the important gaps: what the tool sees, what it cannot see, how deep the search goes, and that output is untrusted. It could be more complete with return-format details, but the provided context is sufficient for correct invocation.

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 schema has 0% description coverage, so the description must compensate. It does convey meaning for several parameters—'tail/search' maps to lines/search, 'surrounding context' maps to context, and paths are implied by 'project log files'—but it never explains the projectPath alias or gives per-parameter semantics. Some inference from parameter names is still required.

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 specifies a distinct operation—'Read-only tail/search of project log files'—and adds scope qualifiers ('surrounding context', 'bounded results') that make it easy to distinguish from siblings like db_query or list_processes. The verb and resource are concrete and unambiguous.

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?

It explicitly says 'Use to inspect errors' and gives clear exclusions: cannot read unrelated terminal stdout unless saved to a file, and search is limited to the last 1 MiB per file. No alternative sibling is named, so it stops just short of full guidance.

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