Loading ToolBench
UGC Brief MCP for creators. Turn a brand brief into deliverables, shot lists, and the right UGC ad format, and catch red flags (unlimited revisions, perpetual usage, spec work, raw-file grabs) before you shoot. The production layer after you land a deal.
Server has 7 well-intentioned tools with clear domain focus (UGC brief analysis). All tools have descriptions (194 - 280 chars, within baseline 34 - 392 range) and input schemas with type definitions. However, critical gaps: (1) Most parameters lack descriptions, only brief_text and format have them; creator_constraints is undescribed. (2) No output schemas documented anywhere, LLMs cannot infer what interpret_brief, scope_deliverables, or build_shot_list return. (3) No error handling guidance, tools fail silently or return unstructured text. (4) Tool names are descriptive but some are compound actions (interpret_brief + catch_red_flags logic bundled). (5) Two tools (brief_checklist, ugc_formats_reference) take no parameters, discovery tools should document what they return. Baseline: 549 production tools average 4 params per tool with 100% param descriptions in A+ tier; this server has ~1.5 params per tool with ~30% described.
Returns the 10-point checklist of everything you must confirm in writing before you shoot: deliverables, usage, exclusivity, revisions, raw files, specs, hooks, talking points, timeline, and brand assets.
Given a UGC format (testimonial, problem-solution, unboxing, demo, GRWM, comparison, hook-test-pack), return a shot-by-shot breakdown: the beats, what to capture, framing notes, and coverage rules to avoid reshoots.
Scan a brief for the 6 most common UGC deal-killers: unlimited revisions, perpetual usage, spec work, raw-file grabs, vague deliverables, and scope creep. Returns the flag, what it means, and how to counter it.
Paste a brand's UGC brief. Returns detected deliverables/specs/usage/disclosure, red flags (unlimited revisions, perpetual usage, spec work, raw-file grabs), and the exact questions to ask before you agree to shoot.
Paste the brief. Returns the best UGC format(s) for the brand's goal, with reasoning: what format wins for this product, audience, and campaign type.
Given a brief, return a structured breakdown: exact video count, format per video, length, aspect ratio, usage scope, revision rounds, and raw-file policy. Confirms what you're actually delivering before you shoot.
No output schemas documented. LLMs cannot infer what interpret_brief, scope_deliverables, build_shot_list, pick_ugc_format, or catch_red_flags return. Without documented return types, agents cannot plan downstream tool calls or extract structured data.
Parameter descriptions missing or incomplete. creator_constraints (scope_deliverables) has no description. brief_checklist and ugc_formats_reference take no parameters but should document what they return. Baseline: 100% of A+ tools have param descriptions.
No error handling guidance. Tools fail silently or return unstructured text. No recovery hints (e.g., 'If brief is ambiguous, ask the user for clarification'). Baseline: A+ tools categorize errors as retryable/user-fixable/fatal and guide next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 46 | 2026-07-28+ | v2 |
Returns the full reference guide to all 7 UGC ad formats: what each is, when to use it, the narrative structure, and the common pitfall to avoid.
Discovery tools (brief_checklist, ugc_formats_reference) lack output documentation. LLMs need to know: what fields does brief_checklist return? Is it an array of {field, question} objects? What structure does ugc_formats_reference use?
build_shot_list format enum is well-defined (7 values), but no guidance on what happens if the brief mentions a format not in the enum. Should the tool suggest the closest match, or error?