flow_skeleton

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

With zero annotations, the description carries the full behavioral burden and largely delivers: it discloses the return semantics (LEADS, not verdicts, to be adjudicated elsewhere), the bounded paginated response ('reports graph counts, never the whole graph'), and the default ordering ('source-rooted first'). This is meaningful behavioral disclosure beyond a bare mutation/read hint.

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?

Dense and front-loaded with the core purpose; every sentence earns its place, including the return-type caveat and pagination bound. It is a long, jargon-heavy wall of text ('nesting-aware {control|sink|lifecycle} streams STITCHED across call seams'), but the complexity of the tool justifies the length and no sentence is filler.

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?

With no output schema, the description correctly compensates by explaining return values (leads, graph counts, never the whole graph), default scoping, and parameter-driven narrowing. It covers everything needed to invoke correctly, though it leaves the relationship to the broader flow family (flow, flow_pass, leads, taint) unclarified — a minor gap against the sibling set.

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 coverage is 60% (kind, format, function are described; limit/offset only have defaults). The description adds real semantics beyond the schema: `function` scopes to one entry and renders skeletons, `kind` filters reach|typestate, and the no-arg case yields every lead, source-rooted. Limit/offset are self-evident pagination with schema defaults, so the modest gap is acceptable.

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 pair ('compose per-function summaries into linear, nesting-aware streams STITCHED across call seams') and explicitly names what it is NOT: the single-function `skeleton`. The two skeleton kinds (REACH, TYPESTATE) are each given a concrete definition with the bug classes they feed (double-free/use-after-free/leak). An agent can distinguish it from its closest sibling `skeleton` on purpose alone.

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 clear routing context: it explicitly contrasts with the single-function `skeleton`, and tells the agent the output is LEADS to be adjudicated with sources_of/reaches, plus the no-arg default behavior. It does not, however, spell out when to pick this over the many other flow siblings (flow, flow_pass, leads, taint) beyond the single-function contrast, so exclusions are implied rather than enumerated.

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