check_cua_actions

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

With no annotations, the description carries the full burden and does well: it states the tool is pure rule-based ('不调 LLM、不耗额度'), advisory-only ('未执行实际拦截'), and discloses that task_scope can only escalate, that optional delegation-chain fields have zero impact when omitted, and that verdicts are aggregated to L3 with a risk_diluted flag on cross-agent propagation. It omits auth/permission requirements and error/limit behavior, so it is strong but not exhaustive.

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?

The content is front-loaded with the core verb, but it is delivered as one very long semicolon-joined paragraph and repeats schema-level details (delegation_depth escalation, task_scope narrowing). Information density is high, yet structure and non-redundancy are only adequate.

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?

There is no output schema, so the description must cover returns, and it does: matched_rules, outcome, per-level counts, ruleset_version, disclaimer, and chain_trace are all named, alongside which rule categories are covered. An agent has enough to call and interpret the tool, with only minor gaps around permissions and failure modes.

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 schema already documents every action field, alias, and nested scope property, making 3 the baseline. The description adds some cross-parameter meaning (scope narrowing only escalates, chain_id/agent_id/parent_agent_id trigger server-side chain taint analysis, session_id writes to the audit table), but it largely restates what the schema descriptions already say.

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 precise verb+resource: grade a Computer-Use Agent action trajectory, returning per-step L0–L3/takeover verdicts plus matched_rules, outcome and summary counts. It clearly distinguishes itself from a generic safety check by being trajectory/action-oriented, but never names its siblings (check_cua_code_audit, check_safety, verify_text) to make the boundary explicit.

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 a concrete usage context: '用于在 agent 执行前/后做安全自检' (run before/after agent execution as a safety self-check). That is clear context, but there are no exclusions and no explicit routing against the sibling safety tools, so it stops short of a full when/when-not statement.

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