JPeetz/agent-skills / skills/api-backend/kie-ai-platform
kie-ai-platform
VerifiedUse this skill when an agent needs to integrate with the Kie.ai API for image/video/audio generation, task management, file uploads, or callbacks. Activates when working with kie.ai models or the platform's asynchronous task API. Covers the full platform: image gen, video gen, music, chat, file upload, credits, webhook verification, and error handling.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 633 B |
| SKILL.md | skill md | 11.0 KB |
| evals/evals.json | config | 3.6 KB |
evals/evals.json
View on GitHub[
{
"id": "kie-001",
"description": "Create async generation task with Bearer auth",
"input": "Generate an image with gpt-image/1.5-image-to-image via Kie.",
"shouldTrigger": true,
"expect": "Agent POSTs to /api/v1/jobs/createTask with Authorization: Bearer header and input as a stringified JSON (json.dumps), gets back a taskId.",
"assertions": [
"uses Authorization: Bearer header, never x-api-key",
"input for gpt-image models is json.dumps() string, not a nested dict",
"HTTP 200 is treated as task creation only, not completion"
]
},
{
"id": "kie-002",
"description": "Poll task result and parse double-encoded resultJson",
"input": "Check whether my image generation task task_abc123 is done and get the result URL.",
"shouldTrigger": true,
"expect": "Agent polls GET /api/v1/jobs/recordInfo?taskId=... and parses resultJson twice: json.loads(data[\"resultJson\"])[\"resultUrls\"].",
"assertions": [
"polls recordInfo endpoint with taskId query param",
"resultJson is parsed twice because it is a nested JSON string",
"result URLs are downloaded immediately since they expire"
]
},
{
"id": "kie-003",
"description": "Verify webhook HMAC signature before trusting callback",
"input": "A Kie callback POST arrived at my server — verify it is authentic.",
"shouldTrigger": true,
"expect": "Agent computes base64(HMAC-SHA256(taskId + \".\" + timestamp, webhookHmacKey)) and compares with X-Webhook-Signature using hmac.compare_digest.",
"assertions": [
"message is taskId + \".\" + x-webhook-timestamp",
"signature verified with hmac.compare_digest against X-Webhook-Signature",
"webhookHmacKey comes from Kie settings, not hardcoded"
]
},
{
"id": "kie-004",
"description": "Wrong auth header pitfall — x-api-key must not be used",
"input": "Set up API access to Kie using the x-api-key header like other providers.",
"shouldTrigger": true,
"expect": "Agent flags that x-api-key is wrong — it returns 401 silently — and uses Authorization: Bearer <key> instead.",
"assertions": [
"x-api-key is rejected as an anti-pattern",
"Authorization: Bearer header is used",
"key is managed at kie.ai/api-key, never hardcoded"
]
},
{
"id": "kie-005",
"description": "File upload via free upload API",
"input": "Upload this reference image to Kie so I can use it as input for image generation.",
"shouldTrigger": true,
"expect": "Agent uploads to https://kieai.redpandaai.co (file-url-upload for remote URLs, file-stream-upload for binary, file-base64-upload for base64 under 10MB) and notes files expire after 24h.",
"assertions": [
"upload target is the kieai.redpandaai.co file upload base URL",
"upload method matches source type (URL/stream/base64)",
"uploaded file expiry (24h) is noted"
]
},
{
"id": "kie-006",
"description": "Check account credits",
"input": "How many credits do I have left on Kie?",
"shouldTrigger": true,
"expect": "Agent calls GET /api/v1/chat/credit and reports the integer balance from data.",
"assertions": [
"calls /api/v1/chat/credit with Bearer auth",
"reports the credit balance to the user"
]
},
{
"id": "kie-007",
"description": "Do not trigger for non-Kie model generation",
"input": "Use foo-bar model from some other vendor's API to make a video.",
"shouldTrigger": false,
"expect": "No Kie platform involvement — skill covers Kie.ai models and its async task API only.",
"assertions": [
"skill stays dormant for non-Kie providers"
]
}
]Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |