civitai-orchestration
Query and explore Civitai Orchestration workflows, jobs, and results. Use for analyzing image/video generation jobs, viewing job results, searching by workflow ID, job ID, user, or date range.
Best use case
civitai-orchestration is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Query and explore Civitai Orchestration workflows, jobs, and results. Use for analyzing image/video generation jobs, viewing job results, searching by workflow ID, job ID, user, or date range.
Teams using civitai-orchestration should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/civitai-orchestration/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How civitai-orchestration Compares
| Feature / Agent | civitai-orchestration | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Query and explore Civitai Orchestration workflows, jobs, and results. Use for analyzing image/video generation jobs, viewing job results, searching by workflow ID, job ID, user, or date range.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
AI Agents for Coding
Browse AI agent skills for coding, debugging, testing, refactoring, code review, and developer workflows across Claude, Cursor, and Codex.
Best AI Skills for Claude
Explore the best AI skills for Claude and Claude Code across coding, research, workflow automation, documentation, and agent operations.
AI Agent for YouTube Script Writing
Find AI agent skills for YouTube script writing, video research, content outlining, and repeatable channel production workflows.
SKILL.md Source
# Civitai Orchestration Interact with the Civitai Orchestration API to query workflows, view job details, and retrieve generation results. ## Quick Start ```bash # Query workflows for a specific user node .claude/skills/civitai-orchestration/civitai-client.js user 12345 # Get a specific workflow by ID node .claude/skills/civitai-orchestration/civitai-client.js workflow <workflowId> # Get job details with scan results (hive_csam_score, etc.) node .claude/skills/civitai-orchestration/civitai-client.js job <jobId> # Get step details from a workflow node .claude/skills/civitai-orchestration/civitai-client.js step <workflowId> <stepName> # Download result images/videos from a workflow node .claude/skills/civitai-orchestration/civitai-client.js results <workflowId> ``` ## Searching & Filtering ### By User ID ```bash # Get workflows for user 12345 node civitai-client.js user 12345 # With options node civitai-client.js user 12345 --take=5 --excludeFailed ``` ### By Workflow ID ```bash # Direct lookup by workflow ID node civitai-client.js workflow 0-019be44b-181e-7a7e-ab1b-b58dc7610dca ``` ### By Date Range **Note:** Date filtering may have limited functionality depending on API access level. ```bash # Workflows from the last 7 days node civitai-client.js workflows --from=2024-01-15 --to=2024-01-22 # Workflows from a specific day node civitai-client.js workflows --from=2024-01-20 --to=2024-01-20 # Dates are assumed to be UTC. Add Z suffix for explicit UTC: node civitai-client.js workflows --from=2024-01-20T06:00:00Z --to=2024-01-20T12:00:00Z ``` **Alternative:** Use `--oldest` to get workflows from oldest first, then paginate: ```bash # Get oldest workflows first node civitai-client.js workflows --oldest --take=50 ``` ### By Tags Tags are set when workflows are created. Common tag patterns: ```bash # Filter by a specific tag node civitai-client.js workflows --tag=user:12345 # Note: Multiple tags can be specified (AND logic) node civitai-client.js workflows --tag=project:myproject --tag=type:image ``` ### By Metadata Search The `--query` option searches workflow metadata: ```bash # Search metadata for a string node civitai-client.js workflows --query="portrait" ``` ### By Status The API supports excluding failed workflows but not filtering by specific status: ```bash # Exclude failed/canceled workflows (get only succeeded/processing) node civitai-client.js workflows --excludeFailed # Include all statuses (default behavior) node civitai-client.js workflows ``` **Note:** To find specific statuses, retrieve workflows and filter client-side, or use metadata/tags when creating workflows for better filtering. ### Pagination ```bash # Get 50 results node civitai-client.js workflows --take=50 # Get next page using cursor from previous response node civitai-client.js workflows --take=20 --cursor=<nextCursor> ``` ### Combined Filters ```bash # Successful workflows for a specific tag, excluding failures node civitai-client.js workflows \ --tag=user:12345 \ --excludeFailed \ --take=50 # Get workflows with metadata search node civitai-client.js workflows \ --query="portrait" \ --excludeFailed \ --take=20 ``` ## Commands Reference ### test Test API connection and verify credentials. ```bash node civitai-client.js test ``` ### user Query workflows for a specific user by their Civitai user ID. ```bash node civitai-client.js user <userId> [options] ``` **Options:** | Option | Description | |--------|-------------| | `--take=<n>` | Number of results (max 10 per API limit) | | `--tag=<tag>` | Filter by tag | | `--query=<text>` | Search workflow metadata | | `--excludeFailed` | Exclude failed/canceled workflows | | `--oldest` | Sort from oldest to newest | ### workflows List and search workflows with optional filters. ```bash node civitai-client.js workflows [options] ``` **Options:** | Option | Description | |--------|-------------| | `--tag=<tag>` | Filter by tag (can specify multiple) | | `--query=<text>` | Search workflow metadata | | `--excludeFailed` | Exclude failed/canceled workflows | | `--oldest` | Sort from oldest to newest | | `--take=<n>` | Number of results (default: 20) | | `--cursor=<cursor>` | Pagination cursor | | `--from=<date>` | Start date (may have limited support) | | `--to=<date>` | End date (may have limited support) | ### workflow Get details of a specific workflow. ```bash node civitai-client.js workflow <workflowId> [--wait] ``` **Options:** | Option | Description | |--------|-------------| | `--wait` | Wait/poll for completion (with timeout) | ### job Get details of a specific job including scan results from event context. This is useful for investigating content moderation results - the `lastEvent.context` contains scan scores like `hive_csam_score` and `hive_vlm_summary`. ```bash node civitai-client.js job <jobId> [--raw] ``` **Options:** | Option | Description | |--------|-------------| | `--raw` | Output raw JSON instead of formatted summary | **Example output:** ``` Job Summary: ID: 58de87d7-d594-4d71-ae43-dd8fc1bcbd23 Type: TextToImageV2 Workflow ID: 8484131-20260121222126332 Prompt Classification: sexual, young, scan Results (2 blob(s)): - JNFDW54JNTATNW42HACYCWSQP0.jpeg (available: true) Last Event: Type: Succeeded Provider: ValdiAI Event Context (scan results, metrics): ** hive_csam_score: 0.00 %, 0.00 % ** hive_vlm_summary: X, No_Child ``` ### step Get details of a specific step within a workflow. ```bash node civitai-client.js step <workflowId> <stepName> ``` ### results Download or view results (images/videos) from a workflow. ```bash node civitai-client.js results <workflowId> [--download] [--dir=<path>] ``` **Options:** | Option | Description | |--------|-------------| | `--download` | Download result files | | `--dir=<path>` | Download directory (default: ./civitai-downloads) | ### blob Get a blob (image/video) by ID with optional NSFW handling. ```bash node civitai-client.js blob <blobId> [--download] [--nsfw] ``` ## Environment Variables Stored in `.claude/skills/civitai-orchestration/.env`: ```env CIVITAI_API_TOKEN=your_bearer_token CIVITAI_API_URL=https://orchestration-new.civitai.com ``` ## API Reference ### Workflow Query Parameters The workflows endpoint supports these query parameters: - `tags` - Array of tags to filter by - `query` - Search workflow metadata - `fromDate` / `toDate` - Date range (ISO 8601) - may have limited support - `cursor` - Pagination cursor - `take` - Number of results (default: 100) - `excludeFailed` - Exclude failed/expired/canceled workflows (boolean) - `ascending` - Sort oldest to newest (boolean) ### Workflow Statuses | Status | Description | |--------|-------------| | `preparing` | Workflow being prepared | | `scheduled` | Scheduled for execution | | `processing` | Currently running | | `succeeded` | Completed successfully | | `failed` | Failed with error | | `canceled` | Was canceled | | `expired` | Timed out | | `deleted` | Deleted | ### Step Types (Recipes) Image/Video generation steps you might encounter: - `textToImage` - Text to image generation - `imageGen` - General image generation - `videoGen` - Video generation - `videoEnhancement` - Video upscaling/enhancement - `videoFrameExtraction` - Extract frames from video - `videoUpscaler` - Upscale video - `videoInterpolation` - Frame interpolation - `convertImage` - Convert image formats - `comfy` - ComfyUI workflow execution ## Examples ### Query a user's recent image generations ```bash # Get user 12345's recent workflows node civitai-client.js user 12345 # Exclude failed jobs node civitai-client.js user 12345 --excludeFailed ``` ### Get workflow and view results ```bash # Get workflow details node civitai-client.js workflow abc123-def456 # View/download results node civitai-client.js results abc123-def456 --download --dir=./my-images ``` ### Paginate through workflows ```bash # First page node civitai-client.js workflows --take=10 # Next page (use cursor from previous response) node civitai-client.js workflows --take=10 --cursor=<nextCursor> ``` ## Error Handling | Error Code | Meaning | |------------|---------| | 401 | Invalid or expired token - check CIVITAI_API_TOKEN | | 404 | Workflow/Job not found | | 429 | Rate limited - wait before retrying | | 422 | Invalid parameters | ## Output The `user` command displays workflow summaries including: - Workflow ID, status, timestamps - Step types (textToImage, videoGen, comfy, etc.) - Prompts for text-to-image generations For raw JSON output, use the workflow command: ```bash node civitai-client.js workflow <workflowId> ```
Related Skills
worktree
Create and manage git worktrees with automatic environment setup. Creates worktrees at ../model-share-<branch>, copies .env, and runs pnpm install.
ux-design
UX design methodology and external consultation. Use when creating user flows, wireframes, interaction patterns, or getting UX feedback. Provides structured frameworks for user-centered design.
retool-query
Run queries against the Retool PostgreSQL database for moderation notes, user notes, and other Retool-managed data. Read-only by default. Use when you need to query the Retool database directly.
redis-inspect
Inspect Redis cache keys, values, and TTLs for debugging. Supports both main cache and system cache. Use for debugging cache issues, checking cached values, and monitoring cache state. Read-only by default.
ralph
Autonomous agent for tackling big projects. Create PRDs with user stories, then run them via the CLI. Sessions persist across restarts with pause/resume and real-time monitoring.
quick-mockups
Create multiple UI design mockups in parallel. Use when asked to create mockups, wireframes, or design variations for a feature. Creates HTML files using Mantine v7 + Tailwind following Civitai's design system.
postgres-query
Run PostgreSQL queries for testing, debugging, and performance analysis. Use when you need to query the database directly, run EXPLAIN ANALYZE, compare query results, or test SQL optimizations. Always uses read-only connections unless explicitly directed otherwise.
opensearch-admin
Inspect and debug OpenSearch clusters — health, index stats, search performance, query profiling, mappings, shards, and thread pools. Read-only admin operations for monitoring and troubleshooting.
mod-actions
Take moderator actions on users - ban, mute, remove content, manage leaderboard eligibility, send DMs. Use when you need to ban a user, mute them, send direct messages, or take other moderation actions.
metabase
Create and manage Metabase questions, dashboards, and public links. Use when the user wants to build metrics dashboards, create saved questions with SQL queries, or share analytics publicly.
meilisearch-admin
Check Meilisearch index status, tasks, health, and settings. Use for debugging search issues, monitoring indexing tasks, and inspecting index configuration. Read-only admin operations.
freshdesk
Interact with Freshdesk support platform - search/view/update tickets, reply to customers, add notes, look up contacts, and manage Knowledge Base articles. Use when you need to manage support tickets, look up customer information, or work with KB content.