asset_quality_gate_run

Analysiscatalog

Verify Asset Quality

Not in the default tool list

This tool lives in the tool catalog. Discover it with scenario_tools_search (or enumerate everything with scenario_tools_list) and run it via scenario_tool_execute_write. To list every tool directly, connect with ?toolsets=full.

Score a Scenario image asset against the Quality Gate — a generic AI-quality check plus brand-brief compliance when a team or project brief is configured. Returns a pass/warn/fail verdict with 0–100 scores and per-dimension reasons and suggestions. Image assets only. Reads the asset's stored verdict first (free); only when no usable verdict exists — or rerun=true — does it run a new analysis, which charges credits and stores the verdict on the asset. Args: - asset_id: required — the image to score - rerun: optional — force a new analysis even when a stored verdict exists (replaces it; use after the brief or sensitivity changed) - sensitivity: optional — 'low', 'medium', or 'high' threshold strictness for a new analysis (defaults to the team/project setting) - dry_run: optional — true returns the credit cost estimate of a new analysis without running or charging - team_id, project_id: required for OAuth callers Returns: source ('stored' or 'new_analysis') plus quality_gate (verdict, overallScore, aiQualityScore, briefComplianceScore, details). Quality Gate is an Enterprise add-on — calls fail when it is not enabled for the team.

idempotentopen-world

Parameters

Name	Type	Required	Description
asset_id	string	✓	Scenario asset ID of the image to score against Quality Gate.
rerun	boolean	—	Force a new analysis even when a stored verdict exists. Re-verifying replaces the stored result — use after the brand brief or sensitivity changed.
sensitivity	enum(low \| medium \| high)	—	Threshold strictness for a new analysis. Overrides the team/project setting for this call; ignored when the stored verdict is returned.
dry_run	boolean	—	Estimate the credit cost of a new analysis without running or charging. Skips the stored-verdict read.
team_id	string	—	Team ID. Required if user belongs to multiple teams.
project_id	string	—	Project ID to scope the operation to.
response_format	enum(json \| markdown)	json	Output format: 'json' for structured data, 'markdown' for human-readable text.

Example Request

JSON

{
  "asset_id": "asset_img001",
  "team_id": "team_abc123",
  "project_id": "proj_xyz789"
}

Example Response

JSON

{
  "source": "stored",
  "asset_id": "asset_img001",
  "quality_gate": {
    "status": "scored",
    "verdict": "warn",
    "overallScore": 61,
    "aiQualityScore": 55,
    "briefComplianceScore": 70,
    "sensitivity": "medium",
    "details": {
      "aiQuality": {
        "score": 55,
        "reasons": [
          "Waxy skin texture on the subject"
        ],
        "suggestions": [
          "Regenerate with a photographic lighting cue"
        ]
      }
    }
  }
}

Common Use Cases

Gate generated images on pass/warn/fail before shipping them
Check brand-brief compliance of an asset against the team or project brief
Re-score an asset after the brief changed (rerun=true)