triz_subsystem_contract

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

With no annotations, the description carries the full burden and does real work: it discloses that verification is exact over explicit assumptions, bounded and nonblocking, that circular-resource-support and source-support-cap results are reported as undecided rather than failed, that max_support_work caps support checking independently of the state cap, and that caller-supplied checked status is never trusted. That last clause is a genuinely valuable trust-boundary disclosure. It stops short of describing side effects, permission needs, or result shape, and several phrases are too opaque to act on.

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 opening clause is appropriately front-loaded with the declared-versus-verified distinction, and most sentences carry information. But the prose is a chain of dense noun phrases ('bounded closure as flat feasibility', 'declared empirical assertion') with no structural separation between what is checked, what is undecided, and what is capped, which slows comprehension more than the length warrants.

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 six-parameter tool with deep nested objects, no annotations, and no output schema, the description covers verification semantics, undecided handling, and one cap but never sketches what a result looks like beyond the word 'undecided', nor how model/box/contract relate. It is more informative than most siblings but leaves the agent guessing about the response and the framing parameters.

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 only 17%, so the description must compensate and it partly does: it explains the declared-vs-verify meaning of 'mode', names the contract content it evaluates (behavior, assumptions, levels, demands, budgets, operating envelope), and explains max_support_work's role and independence from the state cap. But 'box', 'model', and especially 'max_states' receive no treatment at all in the description, leaving three of six parameters undocumented.

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 first sentence names a specific operation, distinguishing a merely declared subsystem summary from one verified against the inner steady-state relation and a declared resource mode, and the 'mode' enum (declared/verify) anchors that. However, the contrast target is described only obliquely ('a declared subsystem summary') rather than naming the sibling tool triz_subsystem_summary, and the surrounding noun stacks ('bounded closure as flat feasibility', 'source support') make the core purpose hard to extract on a first read.

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 implies verification happens under 'verify' mode and that missing data or unsupported modes yield 'undecided', but never states when an agent should call this instead of triz_subsystem_summary, triz_feasibility, or triz_behavior. No prerequisites, no exclusions, no routing rule; the reader must infer the choice from jargon.

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