get_ticket_history

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

Annotations provide readOnlyHint=true, so the description did not need to restate that. The description adds significant behavioral nuance beyond annotations: it reveals that some entries have Field populated with a numeric field ID rather than a name, that OldValue/NewValue for owner/watcher changes are numeric user IDs that cannot be resolved with available tools, and that OldReference/NewReference are unresolved row IDs. This is exactly the kind of trap-avoidance context that makes invocation and response interpretation safe.

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 for the amount of caveat it encodes. The core purpose is front-loaded, and the two special cases are stated efficiently with exact Type names. The only minor cost is a dense run-on sentence for the custom field case; splitting that into two sentences would help scanning. Still, every sentence earns its place and the length is justified by real trap-avoidance 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 list endpoint with 100% schema coverage and no output schema, the description covers what an agent needs: what comes back, how to interpret the common shape, and how to handle the two anomalous shapes. It also names the only mapping assistant (get_queue_fields) and explicitly states that numeric user IDs and row IDs cannot be resolved here. The only gap is a few structural details (e.g. default ordering), but none of that is essential 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 input schema covers 100% of parameters with descriptions: id is a Ticket ID, page and per_page have default/max values, and fields is a comma-separated list. Given that high coverage, the baseline is 3. The description does not add much beyond the schema for the parameters themselves, but it does add context about how the returned Field/OldValue/NewValue fields should be interpreted, which is more behavioral than parameter-level. The schema already carries the parameter documentation weight, so a 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 phrase, 'Get the transaction history for a ticket,' identifying both the resource (ticket) and the precise scope (transaction history). It further specifies that it returns a list of transaction types and even catalogs the structural variants, which distinguishes it cleanly from the immediate sibling get_transaction and from search_tickets without needing to open those 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?

The description strongly implies when to use it: when you need the history/transactions for a single ticket, as opposed to searching tickets or fetching a single transaction. It also gives situational guidance for interpreting the output, e.g. getting owner/watcher changes and custom field changes. It does not explicitly enumerate sibling alternatives or state 'use get_ticket for current state,' but the context is clear for a caller that needs to understand a ticket's audit trail.

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