Orchestrator Next Action

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

Annotations already declare readOnlyHint/idempotentHint/destructiveHint/openWorldHint, so the safety profile is covered. The description adds genuinely useful behavior the annotations cannot express: done=true means the pipeline is finished, and the final submit still requires human approval. It does not mention the skip_options_check prerequisite relationship to run_options_save, which is the one notable behavioral 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?

Front-loaded with the purpose and organized into 'Use when' / alternative / 'Returns' beats, with zero filler prose. The explicit 'Returns: {ok, done, stage, role, attempts_left, instructions}' is mildly redundant given a separate output schema exists, which is the only thing keeping this short of a 5.

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 two-parameter, read-only loop-driving tool with an output schema, the description supplies the essentials: purpose, unattended-use condition, sibling route, and the human-approval caveat on done. Return values need not be explained since an output schema exists; only the skip_options_check/run_options_save ordering precondition is left implicit.

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 app_dir and skip_options_check are already documented, including the default and its meaning. The description contributes nothing about input parameters – its only parameter-adjacent content (stage, role, retry budget) describes outputs, not inputs. Baseline 3 is correct when the schema carries the load.

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 (Return) and resource (the next action for the autonomous driver loop) and enumerates what the action contains: stage, subagent role, retry budget, instructions. It also names the sibling it is not (pipeline_next), so the agent can discriminate without opening any 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?

'Use when: running unattended' gives the triggering condition, and 'For a manual step use pipeline_next' names the explicit alternative and the condition that selects it. Nothing about routing is left to inference.

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