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

Beyond the annotations (readOnlyHint=false, idempotentHint=true, destructiveHint=false), the description discloses that nothing is committed until a separate tool runs, that stale hunk ids and moved branches are refused, a 512 KiB cap, the no-conflict-markers constraint, and the required write access plus 'repo' scope. Those are real behavioral facts an agent cannot infer from structured fields.

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-loads the core action and follows with the mutually exclusive modes, then the commit/auth footer — a logical order with no filler. It is very dense, with parenthetical clauses carrying several constraints each, which borders on packing but never wastes a sentence.

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?

With nine parameters, a nested `choices` object, no enums in the schema, and no output schema, the definition covers the gaps that matter: which params come from the list tool, mutual exclusivity, size limits, binary/deletion edge cases, auth, and the deferred-commit model. An agent has enough to call it correctly on the first attempt.

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 per-field text is already there; the description still adds value by declaring 'pass exactly one of' the three optional payloads — a constraint the schema does not encode — plus a 512 KiB ceiling and the semantics of 'mine' (PR branch) vs 'main' (base branch). It stops short of explaining the hunk-id origin or structure beyond the tool reference.

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 a specific verb and resource — 'Save a resolution for ONE conflicted file of a PR' — and immediately qualifies it as a draft tied to an exact (base_sha, head_sha) pair. That scoping separates it cleanly from the sibling list tool (gluecron_list_pr_conflicts) and the sibling commit tool (gluecron_commit_pr_conflict_resolution), which it names in-line.

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?

Explicitly states the three mutually exclusive input modes and the condition that selects each: `keep` is 'the only option for binary files and files one side deleted', `choices` is for hunk-level server-side application, `content` for a whole resolved file. It also names the follow-up tool (gluecron_commit_pr_conflict_resolution) and the refusal/reload path when the branch has moved.

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