notes

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

With no annotations provided, the description carries the full disclosure burden and does real work: it declares the tool read-only, explains that writes happen through a separate CLI command, and defines the `[orphaned]` marker so the agent can interpret records whose method no longer exists. Remaining gaps are ordering, result limits, and pagination — behavior an agent would want but which does not affect call correctness.

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 whole definition is two dense sentences plus an example, with the note format and read-only constraint front-loaded before the example query. Nothing is padded, though the parenthetical about the CLI and the bracketed notation pack a lot into one sentence, making it slightly harder to scan than a clean multi-sentence layout.

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?

There is no output schema, and the description compensates by spelling out the returned record format and the `[orphaned]` annotation, which is exactly the right thing to spend words on here. For a three-optional-parameter read tool, the only meaningful omissions are result ordering and any cap on returned notes.

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%, so all three parameters are already documented, and the description largely restates them (the `pattern` default and the text/anchor matching rule both appear in the schema). What it adds beyond the schema is modest but real: the worked example shows `pattern` and `path` used together, and the `path[::method]` notation clarifies what an anchor string looks like. That is a baseline-plus contribution, not a compensating one.

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 states a concrete resource — notes attached to files or methods by people/agents — and even renders the record shape `path[::method] time 📝 text`, so an agent knows it is retrieving annotation records, not code. The retrieval verb is only implied ('Read-only', example query) rather than stated outright, and no sibling is named for contrast, but the resource is distinctive enough among the listed siblings (find_callers, locate, outline) to be recognizable.

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?

Usage is conveyed mostly by example: `{"pattern": "refund", "path": "app"}` plus 'without `pattern`: all notes', which tells the agent how to broaden or narrow a query. It also clarifies the read/write split (notes are written via the CLI `callgraph note add`), which is genuinely useful routing information. However, there is no explicit statement of when to prefer this over `search` or `guide`, and no when-not guidance.

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