← 返回 Skills 市场
Seisoai
作者
Legendarylibr
· GitHub ↗
· v1.1.1
2949
总下载
3
收藏
0
当前安装
31
版本数
在 OpenClaw 中安装
/install seiso
功能描述
Unified media generation gateway for agents. Discover tools dynamically, choose API key or x402 auth, invoke image/video/audio/music/3D/training tools, and h...
安全使用建议
This skill is instruction-only and appears to be what it says: a gateway for media-generation tools. Before installing or using it, consider: 1) You will need to provide either an API key or make on-chain x402 payments — never give your wallet private keys or broad credentials to an agent. 2) Prefer scoped, revocable API keys and verify the exact header/location the skill expects (X-API-Key). 3) The skill allows autonomous network calls (normal) — if you enable autonomous agents, set billing/payment guardrails to avoid unexpected charges. 4) Confirm the endpoint domain (https://seisoai.com) and TLS before sending secrets. 5) Verify the agent implementation enforces the SKILL.md's agent-scoped safety checks (tool allowlists, single-agent scope, no recursive orchestration) if you plan to call /api/gateway/agent/* routes. 6) If you need higher assurance, ask the publisher for source code or a signed homepage/repository; absence of install/code files means there is nothing local to audit.
功能分析
Type: OpenClaw Skill
Name: seiso
Version: 1.1.1
The skill bundle provides instructions for an AI agent to interact with the Seisoai media generation API. The `SKILL.md` file contains extensive and explicit 'Agent-Scoped Safety Controls' and 'Reliability Rules' that actively instruct the agent to prevent common attack vectors such as credential escalation, broad data exfiltration ('No broad discovery exfiltration'), unauthorized actions, and recursive orchestration. All API interactions are clearly defined and align with the stated purpose of media generation. There is no evidence of intentional harmful behavior, obfuscation, or instructions for the agent to perform malicious actions.
能力评估
Purpose & Capability
The name/description (media-generation gateway) matches the runtime instructions: discovery endpoints, invoke endpoints, job polling, and tool lists are all about media generation. No unrelated binaries, installs, or extraneous credentials are requested.
Instruction Scope
SKILL.md confines actions to the seisoai API (discovery, invoke, jobs, agent-scoped routes) and includes explicit agent-scoped safety checks. It also documents an x402 payment flow (sign/pay on Base). This is within scope, but pay/ signing steps imply on-chain payments — ensure agents do not have wallet private keys or auto-pay without explicit user consent.
Install Mechanism
Instruction-only skill with no install spec and no code files — the lowest risk install surface. Nothing is downloaded or written to disk by the skill itself.
Credentials
The skill shows how API keys or payment headers are used (X-API-Key or x402), but declares no required env vars. This is proportionate to the purpose; however, users should not provide broad or long-lived secrets unless needed and should prefer scoped API keys and explicit payment approvals.
Persistence & Privilege
always is false and no config paths or system modifications are requested. Model invocation is allowed (normal default). The skill can cause outbound network calls — standard for an API gateway — so review agent autonomy/billing settings before enabling.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install seiso - 安装完成后，直接呼叫该 Skill 的名称或使用
/seiso触发 - 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
v1.1.1
Unified media gateway: discover tools, x402/API key auth, image/video/audio/3D. Agent-scoped safety controls.
v1.0.29
- Simplified and condensed the setup and security sections for easier onboarding.
- General instructions for SEISOAI_WALLET_KEY and dependencies are now more concise.
- Security best practices and advanced usage notes have been removed for brevity.
- No changes to code or functionality; documentation only.
v1.0.28
- Updated wallet key setup instructions to recommend using a secrets manager (1Password, Doppler, etc.) for better security.
- Removed previous guidance suggesting storage in shell profiles or dotfiles.
- Added explicit advice against persisting wallet key on disk in plaintext.
- Provided improved example commands for secure, session-only environment variable injection.
v1.0.27
- Added explicit mention of the required environment variable SEISOAI_WALLET_KEY in the skill metadata.
- No other user-facing changes.
v1.0.26
- Dependency install instructions updated: manual npm install now required with lockfile and --ignore-scripts for security, instead of automatic install on first run.
- Install prerequisites section renamed to "Dependencies (one-time setup)" and expanded for clarity.
- Dependency list in metadata updated to explicitly require both curl and node.
- Security instructions strengthened to recommend not running npm install while the wallet key is set.
- No code changes or runtime logic updates.
v1.0.25
- Added a dedicated security section with best practices for storing and using your wallet key safely.
- Node.js dependencies (`viem`, `@x402/evm`) are now auto-installed on first run—no manual npm install required.
- Clarified that only `curl` is required as a system dependency.
- Updated the metadata to remove `node` as a required binary, reflecting the new dependency handling.
- Revised documentation to emphasize that the private wallet key is never transmitted and should only be used in a dedicated SeisoAI wallet.
v1.0.24
SeisoAI 2.0.0 – Major Upgrade and Payment Flow Refactor
- Added a new scripts directory with x402 payment signer logic (`scripts/x402-sign.mjs` and supporting files).
- Reworked SKILL.md: concisely documents x402 USDC payment flow, with simplified, up-to-date, and stepwise instructions for integrating SeisoAI’s 120+ AI generation tools.
- Updated error handling and tool reference: now includes prices, tool IDs, and improved sample invocations.
- Requirements and dependencies clarified: now needs both curl and Node.js for signature logic.
- Removes redundant/instructional security/policy sections, focusing on technical integration.
- Users must still explicitly approve each payment before AI tool invocation.
v1.0.23
SeisoAI Skill 1.4.0 (Security-focused update)
- Added a detailed Security Profile section to clarify capabilities, install mechanisms, and credential requirements.
- Introduced a Pre-Installation Checklist for verifying host signer compliance, domain security, recipient address, and user-approval settings.
- Clarified compatibility, versioning, and metadata for better transparency.
- Emphasized stricter requirements for payment approval and signer behavior.
- No functional or API changes—documentation and usability improvements only.
v1.0.22
- Clarified and strengthened requirements for host payment signing: per-request user approval is mandatory; auto-signing and batch approvals are explicitly forbidden.
- Added explicit guidance and checklist for host implementers to ensure user is shown payment details and recipient address with no approval caching.
- Updated skill metadata to set autonomous: requires-approval, enforcing payment confirmation even during autonomous skill invocation.
- Added compliance table distinguishing compliant and non-compliant host signer behaviors.
- Included new metadata fields detailing x402 signer requirements (user-approval, auto-sign policy, payto-verification).
- Improved documentation to prevent unsafe host implementations and reduce financial risk for users.
v1.0.21
seiso version 1.0.21
- Added explicit dependency declaration blocks in SKILL.md, specifying required binaries and capabilities (curl, internet, x402-signer)
- Clarified that no credentials or secrets are required/stored by this skill; all authentication relies on per-request x402 signatures
- Significantly expanded and clarified host signer requirements, especially mandatory recipient address verification and strict user approval for payments
- Updated compatibility and metadata fields for accuracy and improved integration checks
- No functional logic changes to the skill—documentation and integration safety improvements only
v1.0.20
- Added `license: Apache-2.0` field and clarified requirements and compatibility.
- Expanded metadata section with structured fields for author, homepage, emoji, required binaries, and x402 payment details.
- No changes to code or functional behavior. Documentation and metadata update only.
v1.0.19
- Added explicit instructions requiring user consent before payment: always display tool name, amount (USDC), network, and payTo address; proceed only after explicit approval.
- Clarified agent execution flow, emphasizing user confirmation and integrating curl command examples for each API call.
- Updated description and metadata for improved tool invocation guidance and compatibility.
- No changes in underlying API endpoints or tool categories.
v1.0.18
**Major update: streamlined docs, added categories/triggers, improved flow, and clarified result handling.**
- Reorganized documentation for clarity—added concise agent execution flow, tool categories, improved endpoint and result field references.
- Added/updated triggers: now includes "flux", "veo", and other generation-related triggers.
- Updated requires block: simplified to just the needed capability (x402_payment_signer), user confirmation, and supported networks/assets.
- Prioritized user payment flow and confirmation UX; made requirements and prompts explicit and prominent.
- Clarified polling logic, result structures, and sample invocations for all major tool types (image, video, audio, 3D, etc.).
- Generalized and condensed the previous, verbose SKILL.md; now easier for both users and integrators to follow.
v1.0.17
Version 2.2.0
- Added support for trigger phrases such as "generate image", "generate video", "seisoai", and more for easier activation.
- Included the "http" tool in the skill definition to enable outbound HTTP requests.
- No changes to core protocol or security; usage and gateway flow remain unchanged.
- Updated metadata to reflect new version and capabilities.
v1.0.16
- Added a `requires` section specifying dependency on the host's x402 payment signing capability for USDC on Base (eip155:8453), with user confirmation required.
- Clarified protocol/capability requirements in SKILL.md to improve integration transparency.
- No changes to endpoints, features, pricing, or tool categories.
v1.0.15
- No code or logic changes in this release.
- Version number updated to 2.0.0 in documentation.
- Clarified user confirmation requirements: agents must display payment details and get explicit approval from the user before signing payments.
- Updated security section to include mandatory user confirmation steps and example prompt.
- No changes to API or payment flow.
v1.0.14
Seiso skill 2.0.0 introduces a major API and protocol redesign focused on gateway-centric x402 paid AI inference.
- All x402-enabled endpoints are now unified under `/api/gateway/*` (previously scattered routes).
- Added public, unauthenticated discovery endpoints for tool listing, prices, and OpenAPI schema.
- New invocation model: all requests are routed through `/api/gateway/invoke/:toolId` or `/api/gateway/invoke` with payment via USDC on Base mainnet.
- Job polling is standardized via `/api/gateway/jobs/:jobId` for async tasks.
- Security and payment signing requirements are now explicitly documented and strictly enforced server-side.
- Example usages, categories, error handling, and pricing workflow are provided for easier integration.
v1.0.13
- Added explicit instruction to always direct payment to the exact wallet address (`payTo`) provided in the 402 challenge.
- Clarified that substitution or overriding of destination payment wallet from local defaults is not allowed.
- Stated destination wallet integrity as a new invariant: signed payment must target the `payTo` in the challenge.
- Minor wording changes: references to "paid inference requests" updated to "paid requests".
v1.0.12
- Added a new "Payment signing boundary" section to clarify secure handling of payment signing.
- Explicitly disallow requesting, storing, or deriving private keys within the skill.
- Require all payments to be signed only by the OpenClaw runtime-managed signer.
- Mandate a fresh sign operation for each payment challenge and specify failure behavior if signer is unavailable.
- No other functional or endpoint changes.
v1.0.11
SeisoAI x402 protocol skill v1.4.0
- Refocused skill to strict protocol-first x402 payment handling without account/session or browser assumptions.
- Switched documentation to emphasize task-specific endpoints (e.g., /api/generate/*, /api/audio/*, etc.).
- Added step-by-step algorithm and strict invariants for deterministic 402 challenge–pay–retry flow.
- Documented updated queue polling contract for generate routes.
- Clarified error handling and common traps for paid and async requests.
- Removed previous agent-scoped and tool discovery instructions to target direct x402-powered inference only.
元数据
常见问题
Seisoai 是什么？
Unified media generation gateway for agents. Discover tools dynamically, choose API key or x402 auth, invoke image/video/audio/music/3D/training tools, and h... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 2949 次。
如何安装 Seisoai？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install seiso」即可一键安装，无需额外配置。
Seisoai 是免费的吗？
是的，Seisoai 完全免费（开源免费），可自由下载、安装和使用。
Seisoai 支持哪些平台？
Seisoai 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 Seisoai？
由 Legendarylibr（@legendarylibr）开发并维护，当前版本 v1.1.1。
推荐 Skills