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

Annotations already declare non-read-only, idempotent, non-destructive, open-world behavior; the description adds the important consent gate, the post-install registration step, and the downstream init call, which are not derivable from annotations. It does not describe failure modes or that installation mutates the remote host beyond device targeting, but the rich annotations plus an output schema make that a minor gap.

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 critical consent rule is front-loaded, but the paragraph then drifts across post-install registration, intent matching, context inheritance, and trigger-keyword advice without clear ordering. The sentence "Select this tool from the user's intent when it is the best match" is tautological filler that adds nothing after the more concrete clarification about not needing to mention specific names.

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 consent-gated remote install with an output schema and full annotation coverage, the description supplies the key things an agent needs: the consent precondition, the preferred alternative, and the follow-up sequence. Only failure handling and what happens to a pre-existing installation are left unstated, which the annotations and output schema largely cover.

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 the schema already documents all seven parameters including confirm and the device targeting fields; the description adds only interpretive guidance about the destination being explicit or inherited from conversation context. The operational parameters (retry_token, workflow_id, workspace_id, ai_session_id) get no added narrative meaning, so the baseline 3 applies.

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 and resource — installing the official Codegraph CLI on the target computer — and immediately scopes it to a consent-gated action. It also distinguishes itself from the generic shell_execute path ("Prefer this over ad-hoc Shell install scripts") and from its sibling codegraph_setup_init_project by ordering the two steps.

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 explicit when-to-use conditions (after explicit user agreement, confirm=true only then), an explicit alternative to avoid (ad-hoc shell install scripts), and a prescribed follow-up sequence (register registry_hint via MCP Gateway when Manager tools exist, then call codegraph_setup_init_project). This is unusually complete routing guidance.

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