toolmoat

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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, so the safety profile is covered. The description adds genuinely new behavioral context beyond that: this view is read-only in the sense that the agent cannot act on pending changes at all, since only the user can accept them, in a terminal. That authority boundary is the kind of thing annotations cannot express.

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 load-bearing: the first defines the payload, the second defines the authority limit. The most decision-relevant information (what is returned) is front-loaded with 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?

An output schema exists, so return-value enumeration is not strictly required, and annotations cover the read-only posture. For a no-arg listing tool this is nearly complete; the only soft spot is that 'waiting for review' is left without any pointer to how review is resolved (beyond 'user, in a terminal').

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?

Zero parameters, so there is no parameter semantics to explain and the baseline of 4 applies. Nothing in the description misleads about inputs.

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 precise resource (MCP servers pinned by the toolmoat proxy) and enumerates exactly what the view surfaces: per-server pinned tool counts and pending tool changes. The subject matter is naturally disjoint from the scan_*/list_rules siblings, so an agent can distinguish it without opening the schema.

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 rather than stated: it never says 'use this first to see what is pinned' or contrasts itself with scan_installed. The one routing hint is negative — that accepting changes is not possible here and must be done by the user in a terminal — which is useful but is an exclusion, not a when-to-use rule.

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, idempotentHint, and non-destructive behavior, so the safety profile is covered. The description adds what information each rule carries (id, severity, category, file kinds, detection, fix), but that overlaps with the output schema, and it says nothing about ordering, completeness, or how the filters affect results.

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?

A single compact sentence that front-loads the verb and resource before enumerating return content. No filler and nothing buried.

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 zero-required-param, read-only listing tool with an output schema and fully described enum parameters, the definition covers enough to call it correctly. The only real gap is that the filtering behavior (category, min_severity) is never acknowledged in prose.

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 both parameters are documented enums, so the schema does the heavy lifting. The description never mentions the category or min_severity filters, so it adds no meaning beyond the schema and leaves the filtering capability to be discovered there.

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 ("Lists the detection rules toolmoat applies") and enumerates the returned fields, which clearly separates it from the scan_* siblings that perform scanning rather than describing rules. It stops short of explicitly naming those siblings, so differentiation is implicit through the resource rather than stated.

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 points to a discovery/reference use case implicitly by saying it lists the rules toolmoat applies, but it never says when to choose this over scan_path, scan_text, or scan_installed, nor does it mention that results can be filtered. Usage must be inferred.

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 readOnly, idempotent, non-destructive and openWorld=false, so the description need not restate safety. It still adds real context: the scan is offline, covers project files of the server working directory, and mirrors the CLI's no-path mode. It does not mention cost/scale implications of the scan, which is the only notable 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?

One dense sentence with the scope front-loaded and the read-only/offline guarantee saved for the end where it lands as a qualifier. The long product enumeration is the only bulk, and it earns its place by defining scan coverage.

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 and rich annotations are present, so the description only needs to establish scope and safety, both of which it does. The one remaining unknown an agent might want—whether results are truncated or how findings are grouped—is partly handled by the max_findings schema field.

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 every parameter (deep, max_findings, min_severity) carries its own description and defaults, so the schema does the heavy lifting. The description adds nothing about how deep or severity thresholds affect results, which is acceptable at full coverage but leaves the baseline of 3.

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 (scans) and a precise resource (the agent setup already on this machine), and enumerates the exact surface covered: Claude Code, Claude Desktop, Cursor, VS Code, Windsurf, Codex, Gemini CLI, plus working-directory project files. The zero-argument, machine-scoped nature makes it clearly distinct from path/text-oriented siblings like scan_path and scan_text.

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 phrase 'like `toolmoat scan` with no paths' plus 'the agent setup already on this machine' implicitly routes the agent here when no specific target is supplied, versus scan_path/scan_text which take targets. It stops short of an explicit exclusion ('use scan_path instead if you have a directory'), but the contrast is easy to infer.

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, idempotentHint, destructiveHint=false and openWorldHint=false, so the safety bar is low; the description nonetheless reinforces it with 'nothing is executed, written or sent anywhere', which is materially useful for a security-scan tool. It adds nothing about cost, timeouts, scan size limits, or how findings are ordered — that last point is only covered incidentally by the schema.

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, zero padding, and the core purpose leads before the usage advice and the safety guarantee. Every sentence carries distinct information.

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-shape explanation is unnecessary, and annotations plus the description together cover scope, safety and defaults. What remains unaddressed is behavioral edge cases — very large directories, remote/URL paths, or how this differs from scanning installed items — which slightly limits completeness.

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 path resolution rules, the deep flag, min_severity enum and max_findings cap are all documented in structured fields; baseline 3 applies. The description only adds meaning for path by listing the artifact types it accepts, not for the three filter/tuning parameters.

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?

Names a specific verb (scans) and resource (one file or directory) and enumerates the scannable artifact types — skill folder, plugin, hook script, MCP server source or config — plus the threat classes detected (prompt injection, hidden text, credential theft, exfiltration). It stops short of distinguishing itself from siblings like scan_text or scan_installed, so an agent must infer that pasted content goes elsewhere.

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?

Gives a clear usage trigger: 'before a skill, plugin or server is installed, enabled or run'. That is real decision context, but there are no exclusions or named alternatives, so the routing versus scan_text/scan_installed is left implicit.

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, openWorldHint=false, idempotentHint and destructiveHint=false, so 'Read-only and offline' largely repeats structured data. 'Without writing it to disk' adds a concrete clarification about no temp artifacts, but nothing is said about cost, limits, or failure behavior despite the tool producing findings.

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 short sentences, front-loaded with the core action and scope; nothing is buried. The 'Read-only and offline' clause is the one piece that duplicates annotations rather than adding information.

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?

Output schema exists so return values need not be explained, and annotations cover the safety profile. However, the description never says what is being scanned for (the findings/severity model is only implied via sibling terms), nor does it connect to list_rules or explain how results are shaped.

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 content, kind, filename, max_findings and min_severity. The description only restates the kind parameter ('Set kind to say what the content is') and adds no meaning about severity thresholds, findings cap, or filename handling.

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 clear verb and resource ('scans content passed inline') and scopes it with examples of acceptable content plus the contrast 'without writing it to disk', which implicitly separates it from path- and install-based siblings. It never names scan_path/scan_installed explicitly, so the differentiation is inferred rather than stated.

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 inline-vs-disk framing implies when to prefer this over scan_path, and 'Set kind to say what the content is' hints at the selection parameter. But there is no explicit when-to-use/when-not rule and no mention of alternatives by name.

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