investigation_log

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

With no annotations, the description carries the full behavioral burden and does substantial work: it declares write semantics, the ordering discipline (command logged BEFORE execution, result afterward), the default anchor, and the one-entry-per-experiment invariant. It omits what happens if no investigation is open and whether entries can be amended or removed.

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?

Front-loaded with the operation type and tightly written; nearly every clause carries a distinct rule. The telegraphic shorthand 'HEAD(+.config)' and the compressed kind list border on cryptic, which costs it a point on clarity despite the efficiency.

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?

Output schema exists, so return values need no explanation, and the description covers the mutation workflow, entry taxonomy, and sequencing rules well for a 6-parameter write tool with no annotations. The unaddressed `refs` parameter and undefined failure behavior (no open investigation) are the remaining gaps.

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 0%, so the description must compensate. It defines `kind`'s valid values, explains `command` and `result` including their enumeration (good/bad/inconclusive), and gives `anchor`'s default. However, `refs` — an opaque anyOf array of free-form objects — is never mentioned, and `text` is left implicit.

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?

Opens with 'WRITE. Append to the open investigation.' — a specific verb, resource, and scope (the *open* investigation) that separates it from read-oriented siblings like investigation_status and from the lifecycle siblings investigation_open/close. An agent can tell immediately this creates log entries rather than reading or closing an investigation.

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 strong conditional guidance: which `kind` values exist, that `ruled_out` requires a one-line reason, and that an experiment needs `command` logged before running with `result` recorded after. It also forbids batching ('one entry per experiment, never a summary of several'). It does not name sibling alternatives (checkpoint, store_memory, investigation_status), so it stops short of a 5.

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