Loading ToolBench
MCP server to control Adobe After Effects through ExtendScript execution, composition management, layer manipulation, and rendering capabilities
Critical gaps across all dimensions. Tool descriptions are present but generic (avg ~40 chars, well below 194-char baseline). Input schemas are severely incomplete: 5 of 8 tools have empty input objects ({}), violating the requirement that every parameter needs a description and type. No output schemas documented. No error handling guidance. Parameter naming lacks consistency (aftereffects_* prefix is redundant; verbs are weak). No security considerations visible. This server would not pass code review.
Add a solid layer to the active composition.
Add a text layer to the active composition.
Apply animation to the selected layer.
Check status of Adobe After Effects and server connection.
Create a new composition in After Effects.
Execute ExtendScript code in After Effects.
Launch After Effects if it is not already running.
5 of 8 tools have empty input schemas ({}). These tools provide no parameter guidance to LLMs.
All tool descriptions are under 50 characters and generic ('Check status...', 'Add a solid layer...'). Current descriptions fail to guide LLM selection.
No output schemas documented. LLMs cannot plan downstream calls or extract required fields. E.g., aftereffects_create_comp should document what fields it returns (comp_id, width, height, duration, etc.).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 22 | 2026-07-28+ | v2 |
Render the active composition.
No error handling or recovery guidance. If aftereffects_execute_script fails, LLM receives no actionable next step. No distinction between retryable vs fatal errors.
Tool names use redundant 'aftereffects_' prefix and weak verbs. 'aftereffects_check_status' → 'get_status'; 'aftereffects_add_solid' → 'add_solid_layer'. Prefix adds no value and wastes tokens.