Trace Runtime Causality

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

Annotations already carry destructiveHint=true, and the description adds meaningful behavior beyond that: the mechanism (combining a stopped snapshot with repeated watchpoint tracing, which implies resuming the target), the bounded stop budget, and the epistemic caveat that an observed writer is not automatic proof of root cause. No contradiction with the annotations; this context is genuinely additive.

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 sentences with the core purpose front-loaded, followed by usage criteria, then exclusions and a result-interpretation caveat. Every sentence earns its place; there is no filler and no repetition of what the schema or annotations already declare.

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 tool with a full output schema, 100% parameter coverage, and annotations, the description covers purpose, mechanism, usage window, hard exclusions, and a caveat about interpreting results. The only minor gap is that side effects beyond 'unsafe-to-resume' are not elaborated, but destructiveHint already flags the safety profile.

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 all five parameters (name, maxDepth, maxStops, sessionId, timeoutMs) are fully documented in the schema. The description adds no per-parameter semantics beyond the schema, which meets the stated baseline of 3 but earns no extra credit.

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 a specific verb-resource pair ('Build a bounded consumer-to-writer evidence chain for one suspicious debugger-visible value') that is more precise than the title. It distinguishes the tool from siblings through its scope (bounded, single value, watchpoint-based writer tracing) and usage context (post-crash or differential evidence), so an agent can tell it apart from debug_find_writer or debug_this_crash without inspecting schemas.

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 description gives an explicit when-to-use trigger ('after crash or differential evidence identifies a value whose runtime producers matter') and explicit exclusions ('Do not use it for frozen dumps or unsafe-to-resume targets'). It stops short of naming which sibling tools to use in those exclusion cases, so it is clear but not fully prescriptive about alternatives.

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