cortex_run_eval

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

Annotations already signal readOnlyHint=false (mutation possible) and openWorldHint=true, so no contradiction with the description. The description adds the useful fact that it returns the same job shape as cortex_run_job. However, it does not disclose actual side effects — running a workflow_eval may trigger LLM calls or create surface items (evidenced by options.create_surface_item) — which matters for a non-read-only operation.

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 dense sentences, front-loaded with the core purpose and the parent-tool relationship. No filler. It stops just short of 5 because it references 'same job shape' rather than describing it, keeping some semantic debt in the reader.

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 an 8-parameter tool with no output schema and 50% schema coverage, the description leaves too much implicit: what the returned job shape actually contains, what side effects occur, and how the criteria/target parameters fit together. Since it defers entirely to cortex_run_job's behavior without describing it, an agent cannot fully reason about invocation or results.

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 50% (options, criteria, input_refs, input_payload are documented; project_id, target_ref, target_type, question are not). The description adds zero parameter-level meaning, so it does not compensate for the undocumented half. For a wrapper tool it would help to explain how target_type, target_ref, and criteria interrelate, but none of that is present.

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: it wraps cortex_run_job for job_kind='workflow_eval' to check whether a run/case/workflow met its criteria (SLA, policy compliance, action-item ownership). It names the parent sibling explicitly, which differentiates it from cortex_run_job. However, it does not distinguish against other cortex_* siblings like cortex_launch or cortex_ask, so differentiation is partial.

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?

Implied usage is reasonably clear: use it to check whether a target met eval criteria, and it identifies cortex_run_job as the underlying tool. But it never states when to call cortex_run_job directly versus this wrapper, and offers no explicit exclusions or alternatives for the eval scenario.

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