← 返回 Skills 市场
plagtech

AI Compute

作者 Plag · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
22
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install ai-compute
功能描述
27-tool AI compute agent — LLM inference, image generation, video generation, text-to-speech, speech-to-text, embeddings, GPU inference, Bittensor decentrali...
使用说明 (SKILL.md)

AI Compute

27 endpoints for managed AI compute — LLM text inference across 11 models, image generation (FLUX, SDXL), video generation, TTS, STT, embeddings, GPU inference via Replicate, decentralized AI via Bittensor, on-chain intelligence, and prepaid compute credits. Each call is a real x402 micropayment ($0.001–$0.50 USDC).

How to call endpoints

bash {baseDir}/scripts/compute.sh METHOD ENDPOINT '{"key":"value"}'

Workflow strategies

Content generation pipeline — use text-inference for copy, image-generation for visuals, text-to-speech to narrate, video-generation for clips. Chain them together for full multimedia output.

RAG / knowledge apps — use embeddings to vectorize documents, then text-inference to answer questions against them. Pair with the deep-research-x402 skill for source retrieval.

Audio transcription — use speech-to-text to transcribe, then text-inference to summarize or extract action items.

On-chain intelligence — use classify-address to profile wallets, classify-tx to categorize transactions, explain-contract to audit smart contracts, summarize for intelligence briefings.

Cost optimization — use compute-futures to prepay credits at a discount, then execute jobs against the balance. Check pricing tiers before large workloads.

Decentralized AI — Bittensor endpoints route inference through the decentralized Bittensor network instead of centralized providers.

Available endpoints (27 tools)

Managed Compute (8 endpoints)

Text Inference — $0.003–$0.10 LLM text inference across 11 models. Supports system prompts, temperature, and max tokens.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute/text-inference '{"model":"llama-3.1-8b","prompt":"Explain zero knowledge proofs simply","max_tokens":500}'

Image Generation — $0.02–$0.08 AI image generation via FLUX and SDXL models.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute/image-generation '{"prompt":"futuristic city skyline at sunset, cyberpunk style","model":"flux"}'

Video Generation — $0.40–$0.50 AI video generation from text prompts.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute/video-generation '{"prompt":"a drone flying over mountain landscape","duration":5}'

Text to Speech — $0.03–$0.05 Convert text to natural speech audio.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute/text-to-speech '{"text":"Welcome to the future of AI compute.","voice":"alloy"}'

Speech to Text — $0.02 Transcribe audio to text.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute/speech-to-text '{"audio_url":"https://example.com/audio.mp3"}'

Embeddings — $0.005 Generate text embeddings for RAG, search, and similarity.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute/embeddings '{"text":"zero knowledge proofs for blockchain scalability"}'

Batch Compute — $0.05 Submit up to 50 compute jobs in one call with 10% discount.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute/batch '{"jobs":[{"type":"text-inference","prompt":"Summarize DeFi"},{"type":"embeddings","text":"DeFi protocols"}]}'

Job Status — $0.001 Poll the status of an async compute job.

bash {baseDir}/scripts/compute.sh GET /api/v1/compute/status/:jobId '{}'

GPU Inference (3 endpoints)

GPU Run — $0.06 Run GPU inference via Replicate. Supports any public Replicate model.

bash {baseDir}/scripts/compute.sh POST /api/v1/gpu/run '{"model":"stability-ai/sdxl","input":{"prompt":"astronaut riding a horse"}}'

GPU Status — $0.005 Poll GPU prediction status.

bash {baseDir}/scripts/compute.sh GET /api/v1/gpu/status/:id '{}'

GPU Models — FREE List available GPU model shortcuts.

bash {baseDir}/scripts/compute.sh GET /api/v1/gpu/models '{}'

AI Chat (2 endpoints)

Chat Completions — $0.04 OpenAI-compatible chat completions endpoint. Multi-turn conversations.

bash {baseDir}/scripts/compute.sh POST /api/v1/chat/completions '{"messages":[{"role":"user","content":"What is restaking?"}]}'

Available Models — $0.001 List available AI models and their capabilities.

bash {baseDir}/scripts/compute.sh GET /api/v1/models '{}'

On-Chain AI Intelligence (4 endpoints)

Classify Address — $0.03 AI-powered wallet classification — identify wallet type, risk level, and activity patterns.

bash {baseDir}/scripts/compute.sh POST /api/v1/inference/classify-address '{"address":"0xd8dA6BF26964aF9D7eEd9e03E53415D37aA96045"}'

Classify Transaction — $0.03 AI-powered transaction classification — categorize tx type, parties, and risk.

bash {baseDir}/scripts/compute.sh POST /api/v1/inference/classify-tx '{"txHash":"0x..."}'

Explain Contract — $0.03 AI-powered smart contract explanation — what the contract does in plain English.

bash {baseDir}/scripts/compute.sh POST /api/v1/inference/explain-contract '{"address":"0xA0b86991c6218b36c1d19D4a2e9Eb0cE3606eB48"}'

Intelligence Briefing — $0.03 AI-generated intelligence briefing on a topic, address, or protocol.

bash {baseDir}/scripts/compute.sh POST /api/v1/inference/summarize '{"topic":"Lido staking protocol"}'

Bittensor Decentralized AI (4 endpoints)

Bittensor Chat — $0.03 Chat completions routed through the decentralized Bittensor network.

bash {baseDir}/scripts/compute.sh POST /bittensor/v1/chat/completions '{"messages":[{"role":"user","content":"Explain subnet mining"}]}'

Bittensor Image Gen — $0.05 Image generation via Bittensor subnets.

bash {baseDir}/scripts/compute.sh POST /bittensor/v1/images/generations '{"prompt":"neural network visualization, abstract art"}'

Bittensor Embeddings — $0.005 Text embeddings via Bittensor.

bash {baseDir}/scripts/compute.sh POST /bittensor/v1/embeddings '{"input":"decentralized machine learning"}'

Bittensor Models — $0.001 List available Bittensor models.

bash {baseDir}/scripts/compute.sh GET /bittensor/v1/models '{}'

Compute Futures — Prepaid Credits (6 endpoints)

Deposit — $0.01 Deposit USDC to open a prepaid compute credit account with tiered discounts.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute-futures/deposit '{"amount":10}'

Check Balance — $0.001 Check remaining credit balance, tier, discount rate, and usage stats.

bash {baseDir}/scripts/compute.sh GET /api/v1/compute-futures/balance '{}'

Execute Job — $0.001 Run a compute job deducted from prepaid balance.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute-futures/execute '{"type":"text-inference","prompt":"Summarize this quarter"}'

Usage History — $0.002 Full usage ledger for your compute credits.

bash {baseDir}/scripts/compute.sh GET /api/v1/compute-futures/history '{}'

Refund — $0.01 Refund unused credit balance to the original depositor.

bash {baseDir}/scripts/compute.sh POST /api/v1/compute-futures/refund '{}'

Pricing — $0.001 View compute futures pricing tiers and discount rates.

bash {baseDir}/scripts/compute.sh GET /api/v1/compute-futures/pricing '{}'

Cost reference

Endpoint Cost Category
GPU Models FREE GPU
Job Status $0.001 Compute
Available Models $0.001 Chat
Bittensor Models $0.001 Bittensor
CF Balance $0.001 Futures
CF Execute $0.001 Futures
CF Pricing $0.001 Futures
CF History $0.002 Futures
Text Inference $0.003–$0.10 Compute
Embeddings $0.005 Compute
Bittensor Embeddings $0.005 Bittensor
GPU Status $0.005 GPU
CF Deposit $0.01 Futures
CF Refund $0.01 Futures
Speech to Text $0.02 Compute
Image Generation $0.02–$0.08 Compute
Classify Address $0.03 Inference
Classify Transaction $0.03 Inference
Explain Contract $0.03 Inference
Intelligence Briefing $0.03 Inference
Bittensor Chat $0.03 Bittensor
TTS $0.03–$0.05 Compute
Chat Completions $0.04 Chat
Batch Compute $0.05 Compute
Bittensor Image Gen $0.05 Bittensor
GPU Run $0.06 GPU
Video Generation $0.40–$0.50 Compute

Compute powered by Replicate, Bittensor, and managed inference infrastructure.

安全使用建议
Install only if you trust the gateway and intend to make paid x402/USDC-backed compute calls. Keep RESEARCH_API_KEY scoped to this service, review each requested endpoint and payload before running it, and require explicit approval for deposits, refunds, prepaid-credit execution, batch jobs, video generation, or any high-cost workflow.
能力标签
cryptofinancial-authorityrequires-walletcan-make-purchasesrequires-paid-servicerequires-sensitive-credentials
能力评估
ℹ Purpose & Capability
The stated purpose and implemented behavior align: it exposes AI compute, on-chain analysis, Bittensor, GPU, and prepaid-credit endpoints through a gateway, and the artifacts disclose that calls are paid x402 micropayments.
ℹ Instruction Scope
The documentation gives concrete endpoint examples, but the helper script accepts arbitrary METHOD and ENDPOINT arguments and does not enforce confirmation, allowlists, or spend limits, so use should remain user-directed.
✓ Install Mechanism
Installation is a normal OpenClaw skill install; metadata declares the required RESEARCH_API_KEY, optional RESEARCH_GATEWAY_URL, and curl/python3 requirements.
ℹ Credentials
Network access and an API/subscription key are proportionate to the gateway purpose, but prompts, payloads, audio URLs, and payment-affecting requests are sent to an external service.
✓ Persistence & Privilege
No local persistence, background execution, privilege escalation, credential-store access, destructive file operations, or obfuscation were found.
如何使用
  1. 确保已安装 OpenClaw（本地或 Docker 部署）
  2. 在对话框中输入安装命令：/install ai-compute
  3. 安装完成后，直接呼叫该 Skill 的名称或使用 /ai-compute 触发
  4. 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of ai-compute-x402: 27-tool AI compute agent with managed and decentralized AI services. - Offers text, image, video generation; text-to-speech, speech-to-text, and embeddings. - Includes endpoints for GPU inference, Bittensor decentralized AI, and on-chain AI analytics. - Add compute futures for prepaid credits with discounted pricing tiers. - All actions are invoked via simple bash scripts with real micropayments per call. - Supports advanced workflows: content generation, RAG/knowledge apps, audio transcription, and blockchain intelligence.
元数据
Slug ai-compute
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

AI Compute 是什么？

27-tool AI compute agent — LLM inference, image generation, video generation, text-to-speech, speech-to-text, embeddings, GPU inference, Bittensor decentrali... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 22 次。

如何安装 AI Compute？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install ai-compute」即可一键安装，无需额外配置。

AI Compute 是免费的吗？

是的，AI Compute 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

AI Compute 支持哪些平台？

AI Compute 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 AI Compute？

由 Plag（@plagtech）开发并维护，当前版本 v1.0.0。

推荐 Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191113 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 留言讨论