invariance_capture_link

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

Given annotations already indicate a non-read-only, non-destructive mutation, the description adds useful behavior: it explains the two modes (setting run_id vs. creating a link), the default target_type, and the standalone nature of captures. It does not contradict annotations and provides context beyond the structured fields, such as the legacy equivalence to invariance_capture_update.

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 two sentences, front-loaded with the core purpose. Each clause adds value: the target types, the two modes, the default, and the context about standalone evidence. There is zero fluff or redundancy, making it efficient for an agent to parse.

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 5 parameters, an output schema is absent, and the schema descriptions cover each field. The description covers the essential decision branches (run_id vs target_id), defaults, and a clarifying note about captures not needing prior execution. It does not explicitly state behavior if both run_id and target_id are provided, nor the return format, but given the openWorldHint and the succinctness, these are minor gaps. The information provided is sufficient for correct invocation in the typical cases.

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 covers all 5 parameters (100% coverage), so baseline is 3. The description adds semantic meaning beyond individual parameter descriptions by explaining the interplay: run_id is a legacy shorthand equivalent to the update tool, while target_id creates a richer link and target_type defaults to 'run'. It also notes link_type defaults server-side. This helps an agent understand how to combine parameters rather than just their isolated meanings.

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 specific verb ('link'), a resource ('capture'), and the domain (evidence-graph target) with explicit target kinds (run/case/workflow_event/node). It distinguishes itself from the sibling invariance_capture_update by noting the legacy equivalence when only run_id is used, and from unlink/links by describing the linking action. This is unambiguous.

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?

Provides clear usage guidance by delineating two parameter-driven modes: 'with only run_id' (legacy PATCH) versus 'with target_id' (richer link), and notes the server-side default for target_type. It also adds context that captures are standalone evidence, aiding the agent's decision of when to use this tool. It does not explicitly exclude alternatives like unlink, but the distinction from the update sibling is clear enough for most use cases.

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