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

Annotations already declare non-read-only, idempotent, non-destructive, non-open-world. The description adds real context beyond them: it is PROJECT-SCOPED (acts only on the explicit project_id, never the active-project pointer), it returns the project identity alongside the result, and the window is 'bit-exact' outside its range, which is safety-relevant. It does not describe error behavior or rate limits.

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 scoping constraint is front-loaded in caps, and subsequent clauses pack parameter and behavioral detail with little waste. The prose is jargon-dense and overloads single sentences (e.g. the ramp/mono drop example), which slightly taxes parsing but each sentence carries information.

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 15-parameter mutation tool with nested objects and no output schema, the description covers most semantics, notes the return of project identity, and discloses the project-scoping contract. The gaps are the unexplained 'stage' enum and the absence of any guidance relative to the many sibling grading tools.

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 only 40% across 15 params, so the description must compensate and largely does: saturation (0 = mono), split_tone's {shadows, highlights, balance}, protect semantics (colours keep their source chroma), ramp_in_f/ramp_out_f easing behavior, and id-to-replace. The 'stage' enum (world/screen) is left completely unexplained, and ev/contrast/temperature/tint are enumerated but undefined.

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 specific verb (grade) and resource (a section of the program spanning [start, end)), and the window/region scoping plus 'bit-exact outside the window' clearly sets it apart conceptually from whole-program grading. It never names the relevant siblings (set_color_grade, set_grade_custom, match_grade, apply_look), so an agent must infer the boundary itself. Clear but not sibling-differentiated by name.

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 implied rather than stated: grade a *section* of the program, and 'Pass id to replace one' hints at the update path. There is no explicit when-to-use vs when-not versus the other grading tools, and no prerequisites or ordering guidance. Adequate but with clear gaps.

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