link_trace

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

With no annotations provided, the description carries full behavioral burden. It discloses undirected traversal semantics, depth cap (<=2 via 'depth <= 2'), and that both limit and depth constrain results. It does not mention cost, performance implications, or what happens when limits are exceeded, though the output schema helps. The 'reference neighborhood' framing plus depth/limit bounds gives reasonable behavioral disclosure.

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 description is compact at two short sentences plus a two-line example block, front-loading the core semantic ('Trace the reference neighborhood of an object'). The examples earn their place by clarifying intent. Minor inefficiency: the depth constraint '<= 2' is repeated in the description and in the depth param default, but overall the structure is clean and readable.

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?

The tool is moderately complex (4 params, 6 seed kinds, bidirectional graph traversal) but benefits from a complete schema and an output schema. The description covers traversal semantics, depth limits, and motivating examples. It does not document edge cases like disconnected seeds or what the output JSON shape contains, but the presence of an output schema mitigates that gap. Adequate for the complexity level.

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 100% and each parameter (kind, id, depth, limit) has a clear description in the schema itself, establishing a baseline of 3. The description adds context around what kinds of ids are valid across entity types ('a uuid for task/report/memo, a run id like wf_cbad7348, a commit sha') via the schema. The description itself adds minimal parameter meaning beyond what the schema already documents.

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 clearly states the tool traces the reference neighborhood of an object using an undirected fanout up to depth 2. It provides concrete example queries ('which tasks/reports touched commit 9d8f020') that make the purpose vivid and unambiguous. This distinguishes it from siblings like unified_search (topic-based) and task_execution_trace (linear execution path).

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 tool purpose implies when to use it (exploration of connected objects around a seed), and the examples clarify intended scenarios like recent-work or commit relationships. However, it does not state when NOT to use it or explicitly name alternative tools, such as unified_search for keyword lookups or memory_search for semantic searches. The guidance is implied rather than explicit.

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