pdb_run

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

With no annotations, the description carries the full burden and does meaningful work: it warns "This can modify the current document" and names hard blocks (arbitrary Python/Scheme evaluation and GIMP shutdown procedures). It does not disclose whether effects are undoable, how failures surface, or whether the operation is atomic, so some behavioral uncertainty remains.

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 action and mutation warning, then moves into encoding details and restrictions in a logical order with no filler. There is minor redundancy between "Search and describe first" and "Inspect the schema first," which convey nearly the same instruction.

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 generic dispatcher with zero schema description coverage, a nested free-form object, and no output schema, the description supplies the encoding contract and safety boundaries an agent needs to construct a valid call. Gaps remain around return values and error behavior, which no output schema covers either.

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% and the nested arguments object is undocumented in the schema, so the description must compensate — and it largely does, explaining that values are JSON scalars/arrays and that object arguments use explicit $ref forms for image, layer, and drawable, plus resolution rules for nested layer indices and selected drawables. The 'name' parameter is only implicitly defined as the PDB procedure name, and no concrete argument example is given.

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: "Run one installed GIMP PDB procedure using named JSON-compatible arguments." The word "one" plus the emphasis on installed procedures makes clear this is a single generic dispatch call rather than a batch facility. It does not explicitly contrast itself with the many purpose-built siblings (scale_image, crop_to_rect, gegl_apply_filter), so the differentiation is only partial.

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?

It gives sequencing guidance — "Search and describe first" and "Inspect the schema first" — which routes the agent toward pdb_search and pdb_describe. However, with ~90 siblings, the crucial question of when to prefer pdb_run over a dedicated wrapper like scale_image or sharpen is never answered, and no exclusion criteria or fallback framing is given.

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