submit_agent_report

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

Annotations already declare a non-read-only, non-destructive, closed-world write, so the safety profile is covered. The description adds behavior the annotations cannot convey: the backend auto-advances the workflow, the response carries workflow/action_required, and the agent should keep working in the same turn. Missing detail on what happens to an already-submitted report or whether it can be resubmitted.

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 compact sentences, with the primary action front-loaded and the advance_project exception clearly separated by parentheses. No filler, though the line-wrapped parenthetical with backend jargon (auto_advanced) is denser than 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?

For a mutating submit tool with an output schema present, the description covers the essential decision logic: an agent knows what is saved and how to continue after the call. The remaining gap is the report object's semantics, which sits in neither the schema nor the description.

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 75%, so project_id and execution_id are already documented with pointer hints (list_projects, get_task_result), and wait_seconds' default and early-return behavior is described in the schema itself. The description adds only loose framing for the report payload ('raw report, context suggestions and host-observed file evidence'), leaving the nested AgentReport fields such as needs_revision, simulated, and tests unexplained. That is near the schema-driven baseline.

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 concrete action plus payload scope: finishing an active role and saving the raw report, context suggestions, and file evidence. It also names the sibling it replaces in the auto-advance case (advance_project), which lets an agent distinguish it without opening schemas. Slightly muddied by the phrase 'ChatGPT role', which is jargon an agent must interpret.

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 conditional rule: when auto_advanced=true the agent must not call advance_project and should follow the returned workflow/action_required in the same turn. That is real routing guidance. It does not cover the inverse case (what to do when auto_advanced is false) or when this tool is inapplicable.

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