Create pipeline

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

Annotations already flag this as a mutating (readOnlyHint=false), non-idempotent operation, so the description isn't burdened with restating that. It adds genuinely useful behavior beyond the annotations: variable values are 'sent only to GitLab and omitted from the normalized response,' and inputs are bounded by depth/length/size — non-obvious constraints an agent should know 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, roughly 50 words, front-loaded with the core action. Each sentence earns its place — purpose, a privacy behavior, and input-type bounds — with no repetition of annotation or schema data.

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 output schema covers return shape, so the description doesn't need to explain that. Given the tool's high complexity (deeply nested inputs schema, 25% parameter coverage), the description covers purpose, types, and limits well but leaves gaps: what ref may be (branch vs tag vs SHA), interaction with pipeline schedules, and the side effect that a new pipeline run is triggered.

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 only 25% — project is the sole parameter with a description. The description compensates for the complex inputs schema by summarizing accepted types (strings, numbers, booleans, arrays, objects) and the depth/length/size bounds, and explains that variables are excluded from the response. However, it leaves ref semantics (branch vs tag vs SHA?) and the variables key/value/variable_type structure to the raw 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 ('Create') and resource ('GitLab pipeline'), further scoped by 'for a ref with bounded structured inputs or variables.' This clearly distinguishes create_pipeline from sibling tools like retry_pipeline, cancel_pipeline, get_pipeline, and especially create_pipeline_schedule, which creates a schedule rather than an ad-hoc pipeline run.

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?

No explicit when-to-use language is present. The phrase 'for a ref with bounded structured inputs or variables' implies the tool is for triggering a new pipeline on a branch/tag/SHA with optional inputs, but it never names alternatives such as retry_pipeline for re-running an existing pipeline or run_pipeline_schedule for scheduled execution.

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