hands_parallel

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

No annotations are provided, so the description carries the full burden, and it delivers: isolated contexts per task, a default concurrency cap of 2, refusal of consequential actions in this mode, shared per-service cooldowns, and per-task plus overall deadlines. It also guarantees cleanup ('contexts always cleaned up'), which is exactly the kind of behavioral detail an agent needs before dispatching parallel work.

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 purpose before the structural detail, and every clause carries operational content — no filler. The rules line is dense and slightly run-on (semicolon-chained clauses), which costs readability but not substance.

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 complex concurrent-execution tool with no annotations and no output schema, the description covers isolation, cleanup, failure policy hooks, deadlines, and the safety boundary. It does not explain what comes back — no output schema exists, so per-task result shape and whether partial failures are reported inline are left undefined, which is a meaningful gap for a fan-out tool.

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 0% (two bare parameters: tasks and max_concurrency), so the description must compensate and largely does — it spells out the task item shape ({id?, actions, on_failure?, deadline_ms?}) and points to hands_batch for the actions sub-schema. The gap is max_concurrency: the schema default is 0 while the description only mentions 'default max 2 concurrent', leaving the 0-means-default mapping and the upper bound unexplained. on_failure and deadline_ms values are also named but not defined.

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 and resource ('Run INDEPENDENT browser tasks on isolated pages') and immediately distinguishes itself from the sibling hands_batch by contrasting parallel versus sequential-with-approval execution. An agent can identify this tool's role without consulting the 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?

It states the trigger condition (tasks must be INDEPENDENT) and an explicit exclusion: 'consequential actions are refused in parallel mode — run them via hands_batch with reflex_ask approval.' That names the alternative and the condition that routes an agent there. The cross-task rule ('NO snapshot refs across tasks, use unique visible text / label / placeholder / name') further tells the agent what will fail.

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