Hostilo

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

Annotations declare this is a non-read-only, non-destructive, non-idempotent create. The description adds real behavioral context beyond them: publishing happens in the same call, the call counts against the plan's site limit, index.html links are rewritten, and missing template fields get sample content. It does not say what happens on a name collision or what the response 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?

Front-loads the outcome and URL before the optional content modes, and every sentence carries information. The middle sentence is dense with nested parentheticals, which costs a little readability but not enough to justify trimming the useful details.

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 creation tool with no output schema and a rich input schema, the description covers modes, constraints, quota impact and template behaviour well. The remaining gap is the return value and error behaviour (e.g. duplicate subdomain), which an agent would need when chaining into get_site or publish_files.

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 baseline is 3, but the description adds meaning the schema does not: the mutual exclusivity of html/files/template, the 'same format as publish_files' contract including the assets/ fallback and link fixing, and what template_fields and template_photos actually do to the output. That is genuine value on top of the field-level descriptions.

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 (creates a new website) plus the resulting URL pattern and the notable side effect that it publishes in the same call. This cleanly separates it from update_site, publish_files, and import_zip, which the agent can rule out without opening their schemas.

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?

Explicitly lays out the three mutually exclusive content modes (html for single-file, files for multi-file, template from list_templates) and tells the agent to use template instead of html/files. It names list_templates as the source of slugs, variants and fields, but gives no guidance on when to prefer create_site over import_zip or update_site.

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 this as a safe, read-only, idempotent operation, and the description adds meaningful behavioral context: Cloudflare verification, exclusion of fake AI bots, and free vs PRO data limits. It does not cover auth requirements or rate limits, but the tier constraints are strong added value.

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 front-loaded with the core purpose and then adds the relevant operator, tier, period, and filtering details in compact form. Every sentence contributes useful information without unnecessary 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?

With no output schema, the description provides enough return-shape context by listing operator names, per-bot counts, pages, formats, countries, and comparison data. It also explains free vs PRO behavior and the fake-bot exclusion, leaving no major gap for an agent to call it 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 both parameters are already documented in the schema. The description reinforces the days options and free-tier limitation, but adds little parameter semantics beyond what the schema already provides, which is the baseline for high-coverage schemas.

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: it shows which AI bots read one site, naming the operators, verification source, and comparison period. This clearly distinguishes it from generic sibling tools like get_site_stats by focusing on AI bot traffic.

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 implies usage through the purpose and tier details, and explains that free accounts get 7 days while PRO supports 7, 30, or 90 days. However, it does not explicitly say when to use this tool instead of get_site_stats or other analytics siblings.

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 read-only/idempotent/non-destructive, and the description adds genuinely new behavior: the 1 MB cap, the explicit TRUNCATED marker with full size, the consequence of republishing a truncated page, and a prompt-injection warning about treating HTML as data. That is exactly the extra context 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?

Three sentences, zero filler, and the primary capability is front-loaded before the size caveat and the safety note. Each sentence carries a distinct, actionable fact.

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 no output schema, the description supplies the return format (HTML or TRUNCATED with full size), the size limit, and the safe follow-up action. An agent has everything needed to call it and handle both normal and truncated responses 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 coverage is 100% and the single "site" parameter is already documented with id/name examples and a pointer to list_sites. The description adds no further parameter meaning, so the baseline of 3 applies.

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 ("Returns the current HTML code of one site") plus the size envelope (up to 1 MB) and the intended use ("so you can edit it"). This clearly separates it from get_site_stats (metrics, not content) and update_site (the write counterpart).

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 use context (fetching HTML in order to edit) and an explicit exclusion — do not send a TRUNCATED response back via update_site. It names the alternative tool, but doesn't state the read-before-write sequencing or contrast against get_site_stats/list_versions, so it falls just short of a full when/when-not map.

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 readOnly/idempotent annotations, it discloses that bots are excluded, no cookies or IPs are used, and that Free accounts lack breakdowns and are limited to 7 days. It doesn't need to explain returns since no output schema exists, but it could note rate limits or freshness. The added context is substantial.

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 three sentences with no wasted words, front-loading the primary metrics before plan-specific details. The list of breakdowns is dense but necessary for completeness, though it 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 read-only stats tool with no output schema, the description fully covers what is returned, plan limitations, period options, and data-collection caveats. An agent has all it needs to call the tool correctly given the plan context.

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 parameter meanings are already fully documented. The description adds plan-based semantics for 'days' (Free accounts always get 7) and implies the site parameter is required by saying 'for one site', but it doesn't give format examples beyond the schema. 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 description opens with a specific verb and resource ('Returns traffic statistics for one site') and then enumerates the exact metrics. It is clearly distinguishable from siblings like list_sites or get_site, which deal with site metadata rather than traffic.

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 implies when to use the tool (to get traffic stats) and states plan-based constraints (Free accounts get only 7 days), but it does not explicitly name an alternative tool or say when not to use it. The context is clear, yet no exclusions are given.

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 (destructive=true, non-idempotent, open world) only flag risk class; the description goes far beyond by disclosing the publishing behavior, assets/ path rewriting, hard limits (20 MB ZIP, 300 files, 10 MB/file, account storage), the allowed file-type whitelist, and the expected_version conflict behavior. This is the level of detail an agent needs before a destructive import.

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 with the core action and outcome before limits and caveats; every sentence carries information (file handling, input methods, quotas, dry-run advice, version history). It is dense but not padded, with only the allowed-types enumeration making it longer than strictly necessary.

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 no output schema, the description fully covers the observable outcome: index.html becomes the published page, other files land in assets/ with folders preserved, relative paths are rewritten, and the prior page is versioned. An agent has enough to call this correctly and predict the effect.

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 baseline is 3, but the description adds real meaning beyond the schema: it explains the two input channels as alternatives, clarifies zip_url accepts public http/https links such as GitHub archives, and reinforces dry_run as a preview step. It adds context rather than merely restating field descriptions.

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 ('Imports a multi-file website from a ZIP archive into an existing site') and immediately distinguishes itself from siblings like create_site and publish_files by operating on an existing site and taking a ZIP archive. An agent can tell what it does and where the result lands 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?

Gives explicit operational guidance ('Use dry_run=true first to see what would happen') and routes the agent to restore_version for recovering the previous index.html. It offers two input channels (zip_base64 or zip_url) but never contrasts this tool with publish_files or create_site, so the sibling-selection reasoning 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 only cover the safety profile (readOnly, non-destructive, idempotent, closed world); the description adds ordering (newest first), scope beyond the caller's own content (sites shared by others, with owner/editor/viewer roles and viewer read-only), and the full pagination contract with has_more/next_cursor. These are behavioral facts the annotations do not convey.

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 with purpose, then scope, then the routing hint, then pagination mechanics. Every sentence carries distinct information and nothing is repeated unnecessarily.

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 no output schema, so the description must cover returns — and it does, listing the fields returned and the has_more/cursor paging signals. For a two-parameter, all-optional read tool with annotations handling safety, nothing an agent needs to call it correctly is 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 coverage is 100%, so the baseline is 3, but the description earns extra credit by explaining the cursor round-trip (take next_cursor from the previous response) and restating the default/max of 100 for limit. That turns an opaque parameter into a usable workflow, though it adds no validation nuance 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?

Names a specific verb and resource (lists websites in the connected Hostilo account) and enumerates the returned fields (live URLs, custom domains, last update time, shared-site roles). It also positions itself against siblings by stating it is the discovery entry point ('Call this first to find a site'), so an agent can distinguish it from get_site/get_site_stats without opening a 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?

The instruction 'Call this first to find a site' gives a clear when-to-use rule and implies get_site is the follow-up. It also states the pagination loop condition (has_more -> cursor = next_cursor), which is actionable guidance. It stops short of explicitly naming when NOT to use it versus a sibling such as get_site for a known site ID.

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, non-destructive, and closed-world, so safety is covered. The description adds genuine context beyond that by disclosing the shape of the return payload (variants, fillable field names, live preview URL) and that results are localized via lang, which matters since there is no output 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?

Two sentences with no filler, and the payoff (what you get) is front-loaded before the follow-up routing sentence. Slightly dense but every clause earns its place.

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, zero-required-param list tool with no output schema, the description covers purpose, return contents, and the natural follow-up call. Nothing essential is missing; only minor items like whether results are cached or paginated are unaddressed, and those are unlikely to matter here.

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 the single lang parameter is fully documented with its enum and default in the schema itself. The description references lang only as a token passed to create_site, adding no format or behavior detail beyond what the schema already provides, so the baseline of 3 applies.

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 ready-made Hostilo site designs') and enumerates exactly what comes back: slug, name, description, variants, fillable fields, and a preview URL. That level of detail clearly separates it from siblings like list_sites or get_site, which deal with user-created sites rather than templates.

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?

Explicitly names the downstream action and the parameters it feeds: 'Use create_site with template, template_variant, template_fields and lang to create a site from one.' This gives the agent a clear reason to call it, though it doesn't state exclusions against alternatives (e.g., when a user wants existing sites instead, use list_sites).

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 profile is covered. The description adds genuinely useful behavioral context beyond that: the result is truncated at 10 entries and ordered newest first, so the agent knows it may not see the full history.

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 well-formed sentence that front-loads the action and resource, then layers the cap, ordering, and returned key with zero filler. Nothing is wasted or 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?

With no output schema, the description carries the return burden and does state the count limit, ordering, and the identifying field. It stops short of describing what each version entry contains (timestamp, name, etc.), a minor gap for an otherwise complete definition.

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 the single 'site' parameter is fully documented in the schema, including format examples and a pointer to list_sites. The description adds no parameter meaning beyond that, so the baseline 3 applies.

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 saved previous versions of a site') and adds scope detail: cap of 10, newest-first ordering, and the version_id each entry carries. An agent can distinguish this read-only inventory tool from the sibling restore_version mutation without opening either 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 only implied — mentioning 'with their version_id' hints that this is the lookup step before restore_version, but the description never states when to use it or names restore_version as the follow-up alternative. No exclusions or prerequisites are given.

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 destructiveHint=true and idempotentHint=true, and the description goes well beyond them: previous index.html is kept in history, other files land in assets/ with folders preserved, and relative links in index.html are rewritten automatically. It also discloses hard limits (300 files, 10 MB/file, 20 MB total, account storage) and the allowed MIME set, which no annotation conveys.

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 with purpose and the create_site alternative, then flows into file placement, encoding, limits and dry_run with almost no filler. The long allowed-extensions enumeration is the one stretch that could be trimmed or left to validation errors.

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 multi-file publish with no output schema, the description covers placement, history retention, link rewriting, encoding, limits, and preview. The only omissions are the destructive clear_assets flag and the expected_version conflict workflow, both of which are documented only in the 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?

Schema coverage is 100%, so the baseline is 3, but the description adds real semantics beyond the schema: the content-vs-encoding rule for binaries ('content is base64 and encoding is "base64"') and the path-rewriting contract that lets callers write normal relative paths. It does not add anything for clear_assets or expected_version, which live only 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?

Specific verb ('publishes') plus resource ('several files at once to an existing site'), and it explicitly distinguishes itself from create_site for new sites. The index.html vs assets/ split is spelled out, so the agent knows the tool's exact scope 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?

Explicit routing: 'use create_site first for a new one', and 'Omit index.html to only add or replace files' tells the agent a meaningful alternate invocation mode. It also names dry_run=true for preview, which is a clear when-to-use signal before a destructive publish.

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 destructiveHint=true and idempotentHint=false. The description adds meaningful context beyond those annotations: the current version is saved to history first, so the restore is undoable. It does not mention auth or rate limits, so it stops 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 with zero waste. The primary action is front-loaded, followed by the important undo 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?

No output schema exists, so return values need not be explained. For this two-parameter, non-idempotent, destructive operation, the description covers the action, the prerequisite source for version_id, the publish side effect, and the undo behavior.

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 both parameters are fully documented in the schema. The description only repeats that version_id comes from list_versions and does not add syntax or format details beyond that.

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: 'Restores a previous version of a site and publishes it.' The parenthetical '(version_id from list_versions)' names a sibling as the source, helping the agent distinguish this from list_versions and update_site.

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?

Implies a prerequisite by telling the agent to get version_id from list_versions, but does not explicitly state when to choose restore_version over alternatives like update_site or publish_files, nor any exclusions.

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 cover the safety profile (readOnlyHint=false, destructiveHint=false, idempotentHint=true), and the description goes well beyond them: the write is immediate and public, the previous version is retained in a 10-entry history and can be restored, and a stale expected_version causes a no-op plus version_conflict rather than a clobber. That is exactly the kind of mutation consequence an agent needs before calling.

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, each load-bearing: the write scope, the reversibility guarantee, and the optimistic-concurrency rule. The most important fact (immediate publish) is front-loaded, and there is 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?

For a mutation tool with no output schema, the description covers consequences, reversibility, history depth and conflict handling, so an agent can call it safely. It omits any note on permissions/auth requirements and whether other site files are touched, which are minor gaps given the annotations and full schema coverage.

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 html, site and expected_version are already documented in the schema, including the version_conflict behavior. The description restates the expected_version contract and adds the merge hint ('get the site again and merge'), which is a small increment over structured data rather than new semantics. Baseline 3 applies.

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 verb and resource plus the exact scope: it replaces the site's index.html and publishes immediately. That is far more precise than the title 'Publish site HTML'. It never differentiates itself from the sibling publish_files, which sounds like it may also write site content, so it stops short of a 5.

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 gives a concrete usage condition for the concurrency case: pass expected_version when several people edit the site, sourced from get_site. It does not, however, say when to prefer this tool over publish_files or import_zip, so the alternative-selection guidance is incomplete.

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