describe_sources

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 discloses the compute profile (fast = no download, deep = local Qwen3.5 2B MLX), the important side effect that results are written into the Project panel's Comment and Log Note columns, and that allowOriginal trades machine load for coverage. Remaining gaps are the async job lifecycle (only 'running' is hinted via the schema and the write is not described as idempotent or undoable), so it falls short of a 5.

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 its output fields, then mode semantics, then the write side effect — a sensible order with no filler. It is dense and packs many clauses into one paragraph, which costs a little readability but every 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 6-parameter, no-annotation, no-output-schema tool, the description supplies the informal return shape (the list of per-shot attributes) and the mutation semantics of writeToProject. The main missing piece is how the call behaves when it exceeds waitSeconds and how to resume or collect results afterwards, which matters for a long-running analysis job.

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 baseline is 3; the description goes beyond by explaining what 'deep' actually adds (a local vision model producing natural sentences) and precisely what writeToProject writes and where (Comment column plus Log Note). It does not add anything for bin, sourceFolder, waitSeconds or allowOriginal beyond the schema's own text.

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 — describe every shot (file) in a footage folder — and enumerates exactly what the analysis produces (shot size, setting, labels, sharpness, exposure, shake, faces, beauty score). This is clearly distinct from siblings such as find_shots, detect_scene_edits or transcribe_to_captions, so an agent can route without opening the schema.

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 usable selection guidance between the two modes ('fast' needs no download, 'deep' adds a local vision model) and notes the extra cost of allowOriginal, which implies when to pick each. However, there is no explicit when-to-use/when-not guidance at the tool level and no naming of alternatives among the many analysis siblings, leaving the agent to infer that this is the per-clip descriptive QC pass.

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