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

With no annotations present the description carries the full burden and does so well: it states the tool never signs or holds anything, that the output is an unsigned tx to be signed and sent by the caller, that signal costs only gas, that bounty funds sit in escrow, and that cancel refunds only after 24 h with no answer. It also discloses the simulation affordance via the optional 'from' address.

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 two facts an agent needs first (unsigned transaction, own wallet), and the action list is compressed into a single semicolon-delimited sentence with no filler. It is dense and reads as one long block, which costs a little scanability, but every clause carries information.

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 9-parameter, no-annotation, no-output-schema tool, the description covers the return payload, custody model, per-action inputs and the escrow/refund lifecycle. Remaining gaps are minor: no mention of what award does to the escrowed bounty timing or how failures surface, but nothing an agent needs to select and invoke correctly is absent.

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 the baseline is 3, but the description adds real meaning the schema does not: the 0.02 USDC self-answer threshold for APEX Faucet, the semantics of target=0 ('anyone') and passport=0, and the fact that 'from' triggers pre-sign simulation. These are routing details that go beyond the field descriptions.

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 ('Talk on G2 on Arc from your own wallet') and immediately discloses the return shape ('returns an UNSIGNED transaction (to, data, value, chainId 5042)'), then enumerates the six actions (signal, ask, answer, award, cancel, claim) with one-line semantics each. An agent can distinguish this from g2_read and g2_jobs_post 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?

Per-action guidance is genuinely decision-grade: it says which action takes a bounty (ask), what target 0 vs a passport means, and that APEX Faucet self-answers for bounties >= 0.02 USDC. It also routes queryId provenance to the sibling g2_read. It stops short of an explicit 'use g2_read to read / do not use this to sign' exclusion, so a 4 rather than a 5.

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