Read Service Dependencies

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

The annotations already declare it as a safe, read-only, open-world read, so the description is not burdened with the safety story. It adds real behavior the annotations do not cover: each edge is reported with rate, failure share and p95 duration, and when no calls are observed it silently falls back to compose depends_on and environment host values. That fallback is exactly the kind of provenance detail an agent needs to interpret results, though the description does not say how the agent can tell which source produced the answer.

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?

Three tight sentences: what is returned and in what metrics, then the fallback condition, then the serviceName clarification. It is front-loaded and free of filler, though the fallback sentence is dense enough to need a second read.

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 usefully enumerates the returned fields (rate, failure share, p95) and explains the dual data source, which is the key interpretive context. The remaining gap is the meaning and effect of env, region and the slug parameters, which are undocumented at 17% schema coverage.

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 carry meaning for six parameters and it only partly does. It gives a genuinely useful clarification that serviceName refers to a compose service (not an app or host slug) and ties results to the range's traces, but env, region, appSlug and workspaceSlug are left undifferentiated in both schema and description.

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 uses a specific verb and resource: it returns the calls observed between a service and other services, databases and hosts. That is squarely about dependency topology rather than raw traces, which implicitly separates it from searchServiceTraces and readTrace. It stops short of naming those siblings, so it does not quite reach the top mark.

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?

The usage context is implied by the trace-derived aggregation and the compose-file fallback, so an agent can infer it is the tool for dependency inventory. However, there is no explicit instruction on when to choose this over searchServiceTraces, readTrace or getAppTracing, and no stated prerequisites such as needing tracing enabled.

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