router_session_push

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

With no annotations provided, the description carries the full behavioral burden and discharges it exceptionally well. It discloses fail-closed validation (rejects bad formats), idempotency (duplicate sessionId pushes are rejected), local-only persistence constrained by a data-sovereignty rule, side effects on a cost ledger, and HMAC chaining for audit. This exceeds what most tool definitions reveal about mutation, durability, and security posture.

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?

Every clause earns its place; the two sentences pack validation, expansion, persistence, sovereignty, idempotency, cost, and audit into a flowing pipeline, and the core purpose is front-loaded. The dense arrow-and-parenthesis structure ('（数据主权铁律——记录不出企业边界...）') and the cryptic '(T7 第七章)' flourish reduce scannability, but there is zero fluff and no repetition of schema content.

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 tool with no output schema and no annotations, the description covers the input contract, failure posture, side effects, idempotency, cost behavior, audit, and continuation-match criteria. The main gap is the response contract — an agent is not told what it receives back on success or on rejection (e.g., error codes or receipt identifiers). The five-tuple sentence is also terse enough that its exact matching rule remains somewhat opaque.

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?

The schema covers the single 'raw' parameter fully (100%), listing its fields, so the baseline is 3. The description adds meaning beyond the schema by revealing semantics: sessionId is the idempotency key, the payload must conform to RouterSessionSchema or be rejected, and the five-tuple (executor, employee identity, model, working directory, runtime) drives the continuation matching logic. This behavioral binding of schema fields to tool policies is genuinely additive.

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 the verb+resource pair ('router 过站 session 承接') and enumerates a concrete pipeline: schema validation, multi-round expansion, desensitized local persistence, cost ledger entry, and HMAC audit chaining. This makes the tool's role as a router session handoff/push clear. However, it relies on heavy domain jargon ('过站', '五元组', 'T7 第七章') and never explicitly contrasts itself with sibling data tools like data_push or device_data_push, so a fresh agent must infer its distinct niche.

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?

No when-to-use or when-not-to-use guidance is given, and no alternative tools are named despite a sibling list containing several data-push and workflow-submit tools that could plausibly compete for the same call. The pipeline description implies the intent (push a router session through this defined flow), but nothing tells an agent how to choose this over data_push, commons_invoke, or workflow_submit. Context is conveyed only implicitly through the behavioral chain.

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