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

With no annotations, the description does carry the main behavioral fact: the call is effectively idempotent and will return an existing record rather than duplicating. It says nothing about required permissions, what happens on invalid email, or partial/match semantics — gaps that matter for a write tool with zero annotation coverage.

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 short sentences, zero filler, with the outbound behavior and the workflow ordering both front-loaded. Nothing could be cut without losing 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 needn't be explained, and the sibling reference covers workflow placement. But for a 2-parameter write tool with no annotations and no schema descriptions, the absence of any parameter or failure-mode detail leaves the definition thin.

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 and does not: neither email_address nor full_name is mentioned, and it never states that email is the identity key or that full_name is optional. The only hint is the word 'sender', which loosely implies email.

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 ('Create a contact') and adds a key behavioral qualifier ('or return the existing one'), which signals an upsert rather than a plain insert. It does not explicitly name find_contact as the counterpart, so the differentiation from that sibling is only implied.

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 precondition: 'Needed before escalate_to_human when the sender is not yet in the CRM.' That tells the agent both the trigger and the ordering constraint. It stops short of saying 'use find_contact instead if the sender already exists', leaving the when-not case to inference.

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?

With no annotations, the description carries the full burden and does it well: it discloses body escaping, appended operator signature, the draft-by-default gate, the degraded send=True behavior when sending is disabled, and the refusal conditions (automated/no-reply senders, already-processed emails). These are exactly the non-obvious behavioral traits an agent needs before invoking a mutation.

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?

Five tight sentences, front-loaded with purpose, then format, then side-effect gating, then failure conditions. Nothing is repeated from the schema or wasted.

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 no explanation, and the description covers the remaining risk surface: side effects, degradation under operator policy, and refusal cases. The only residual gap is the identifier format, which is minor for a tool whose behavior is otherwise fully disclosed.

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, and it does for `body` (plain text, blank-line paragraphs, HTML-escaped, signature appended) and `send` (default false, conditional send). `original_email_id` is only implied by "reply to an incoming email" — its expected format and where to obtain it are never stated.

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 opening sentence gives a specific verb and resource ("Create a reply to an incoming email") that no sibling tool duplicates. The rest of the description sharpens what kind of reply it is (draft by default, sent only under operator policy). It does not name a sibling or alternative path, so an agent gets no explicit routing contrast.

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?

Strong conditional guidance: send=False (default) saves a draft, send=True sends only if the operator enabled sending, and certain senders/emails are refused outright. That tells the agent when the tool will and won't act. It never states when to prefer escalate_to_human instead of replying, which is the one real usage decision left to inference.

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?

With no annotations, the description carries the load and does well: it discloses the three side effects (Case creation, email linkage, marking the email processed), the contact lookup fallback when contact_id is omitted, and the prerequisite of calling create_contact. It does not state reversibility, permission requirements, or what assign_to_team_id's absence implies.

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 tightly packed sentences, front-loaded with the action and followed by the trigger condition and the parameter notes; nothing is redundant. The mid-sentence line break in 'link the email to it / and mark the email processed' is a minor formatting blemish.

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 explained, and the description adequately covers the mutation's effects and the contact-resolution path. The remaining gap is the undocumented `assign_to_team_id`, which an agent must guess at in an escalation flow.

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, and it only partly does: `reason` gets a content instruction ('should say why') and `contact_id` gets fallback behavior plus a prerequisite. `email_id` and `assign_to_team_id` are left entirely undefined beyond their titles, so half the parameters remain ambiguous.

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 compound action (create a Case with origin Email, link the email, mark it processed), which is a distinct behavior no sibling performs — create_reply and mark_processed do parts of it separately. The agent can identify the tool's identity 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 concrete triggering conditions: no grounded KB answer, or topics billing/legal/data loss/security, or an upset customer — a genuinely usable decision rule. It stops short of naming the alternatives it preempts (search_knowledge, create_reply) or stating a when-not case, so usage is clear but not fully routable.

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?

With no annotations, the description carries the full behavioral burden, yet it never says whether this is a read-only lookup, whether it returns zero, one, or many records, or what happens when no contact matches. The one piece of real behavioral context is the dual-field matching rule, which is genuinely useful but not enough on its own.

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 sentence with no filler, and the dual-field matching detail is placed inline where it is immediately useful rather than buried after boilerplate.

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 explained, and the single-parameter surface is simple. However, for a lookup tool the missing pieces — match strictness and no-result behavior — are the ones an agent most needs before invoking it.

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 0% — the parameter is documented only as "Email Address" — so the description must compensate, and it partially does by revealing that the value is matched against either email field. It still omits match semantics: exact vs partial, case sensitivity, and whether a non-matching value is an error or an empty result.

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 ("Find contact records") and adds the lookup key semantics — matching against emailaddress1 or emailaddress2 — which is more precise than the name alone. It does not explicitly differentiate itself from siblings like create_contact or get_email, but the find-vs-create distinction is 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?

There is no statement of when to use this instead of create_contact or get_email, and no prerequisites or exclusions. The agent must infer the use case entirely from the verb.

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?

With no annotations, the description carries the full load and does so for the most important trait: it flags the body as UNTRUSTED third-party text and instructs the agent to treat it as data, which is a genuine prompt-injection guardrail. It stops short of stating whether reading has side effects (e.g., whether it marks the email processed) or any permission/rate constraints.

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 highest-risk information (untrusted body) is placed after a front-loaded purpose statement so it reads naturally without burying the point.

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 details needn't be spelled out, and the safety warning covers the biggest behavioral unknown. The remaining gap is operational: how to source email_id and whether the read mutates state, neither of which is addressed.

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 0%, so email_id is undocumented in structured fields, and the description does not explain its format or where to obtain it (presumably from list_new_emails). It at least makes clear the tool operates on one specific email, but does not compensate for the coverage gap.

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 ("Read one email") and even enumerates what comes back: sender, plain-text body, sender's contact id, and last few emails on the same case. It distinguishes itself from list_new_emails by being a single-item read, though it never names a sibling explicitly.

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: an agent can infer this is the tool for fetching one email when it already has an id. There is no statement of when to prefer it over list_new_emails or find_contact, and no prerequisites or 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?

With no annotations, the description carries the load and does well: it discloses the non-obvious side effect that automated senders are skipped and auto-marked processed, that results are oldest-first, and that only metadata is returned. It omits rate limits or whether listed emails themselves get marked processed, keeping it below 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?

Three tight sentences: purpose first, return shape second, side-effect caveat third. No filler, and the important behavioral detail is front-loaded.

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 the return shape needn't be described, and the description still adds useful semantics (auto_skipped). The main omission is the meaning of 'top' and any paging behavior, which an agent needs 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?

The single 'top' parameter is undocumented in the schema (0% coverage) and never mentioned in the description, so an agent gets no guidance on the page size, its default of 10, or how to retrieve more results. This is the clear gap in an otherwise strong definition.

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 plus the defining filter (unprocessed) and ordering (oldest first), which cleanly separates it from get_email and the other siblings. An agent can tell exactly what this returns.

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 routes the agent to get_email for message bodies, which is the key next-step decision. It gives clear context for use but does not name when-not-to-use cases (e.g., after mark_processed or when paging).

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?

With no annotations, the description carries the full burden. It usefully discloses that no reply is sent and that the target is already-handled mail, but it never says what 'processed' actually does (archive, move, label?), whether it is reversible, or what side effects occur. The output schema covers return shape, so that omission is less costly.

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 tight sentence with the core constraint front-loaded and examples packed into a parenthetical. No redundant restatement of the tool name.

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 one-parameter mutation with an output schema, the purpose and usage hints are largely sufficient. The gap is functional ambiguity about what marking as processed changes on the mailbox, which matters for an irreversible write.

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 needed to explain the single required parameter, and it does not mention email_id at all. An agent must infer that the argument is an email identifier rather than, say, a subject line or index.

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 ('mark ... as handled') plus resource ('an email') and a clarifying constraint ('without replying') that separates it from create_reply. The parenthetical examples of spam/thank-you/duplicates sharpen the intent further. It stops short of naming a sibling explicitly, so it is clear but not fully differentiated.

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?

'Without replying' plus the example categories imply when this is appropriate versus create_reply or escalate_to_human, but the routing is inferred rather than stated. No explicit when-not guidance 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?

With no annotations, the description must carry the full burden. It does disclose two meaningful behavioral traits: only published articles are searched, and grounding/escalation is required when results are empty. However, it omits result limits/pagination, whether search is keyword vs semantic, and any permission or rate-limit context.

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 purpose, then return shape, then usage policy. The return-fields sentence partially duplicates the existing output schema, which is the only mild 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?

Purpose, grounding policy, and escalation fallback are all covered, and an output schema exists so return values need little explanation. The notable gap is the undocumented 'top' parameter and the absence of any note on result limits or empty-result handling details.

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 explain both parameters. It adds meaning for 'query' (keywords extracted from the customer's problem, not a full question), but 'top' is never mentioned, leaving the result-count control entirely unexplained.

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 ('Search published Knowledge Articles') plus the input source ('keywords from the customer's problem'). Sibling tools are all email/contact/escalation operations, so there is no ambiguity about which tool retrieves knowledge.

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 an explicit usage policy: answer only from results, and if nothing relevant is returned, escalate rather than guess — which names the fallback action matching the escalate_to_human sibling. Both the when and the when-not are stated.

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