← 返回 Skills 市场
🔌

TACE

作者 Tender · GitHub ↗ · v1.0.6 · MIT-0
cross-platform ✓ 安全检测通过
116
总下载
0
收藏
0
当前安装
7
版本数
在 OpenClaw 中安装
/install tace
功能描述
TACE (Tender Agentic Commerce Engine) runtime contract. Includes concrete payloads required by current endpoints.
安全使用建议
This skill appears to be what it says: an API contract for TACE. Before installing, verify the production base URL and GitHub source link, and confirm you will use a proper wallet signing flow (do NOT paste private keys or secret seeds into the agent). Treat the listed dev base URL as development only, and ensure any webhook/callback URLs you provide are under your control to avoid unintended data exposure.
功能分析
Type: OpenClaw Skill Name: tace Version: 1.0.6 The TACE (Tender Agentic Commerce Engine) skill bundle provides a structured API contract for an AI agent to perform commerce-related tasks such as authentication, product discovery, and order management. While it handles sensitive personal information (PII) for shipping and registration in skill.md, this behavior is explicitly documented and aligned with its stated purpose as a commerce engine. The skill includes strong security guidelines for the agent, such as redacting sensitive values and avoiding arbitrary command execution, and shows no signs of data exfiltration, obfuscation, or malicious intent.
能力标签
cryptorequires-walletcan-make-purchasesrequires-sensitive-credentials
能力评估
Purpose & Capability
The name/description (TACE runtime contract) matches the SKILL.md: it enumerates endpoints, request bodies, and auth flows. Required binary (curl) and optional env vars (TACE_ENV, TACE_SKILL_VERSION) are proportional to an HTTP-based integration.
Instruction Scope
Instructions are limited to composing and calling the listed API endpoints and handling retries/verification. The auth flow requires EIP-191 signing (a wallet-based signature) which is appropriate for the described API; the contract explicitly forbids requesting or storing confidential user material.
Install Mechanism
There is no install spec and no code files — instruction-only — so nothing is downloaded or written to disk by the skill itself.
Credentials
No credentials or sensitive environment variables are required; the two declared env vars are optional and relate to environment/versioning. There is no request for unrelated tokens or keys.
Persistence & Privilege
The skill is not always-enabled and allows user invocation; it does not request persistent presence or modify other skills or system settings.
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 tace
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /tace 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
v1.0.6
**Expanded to include concrete endpoint payloads and response schemas.** - Documents required request bodies, query parameters, and responses for all endpoints. - Adds explicit contract for authentication payloads (nonce, verify, register). - Moves "environment" configuration to TACE_ENV and updates default base URL. - Clarifies catalog, ordering, subscription, and feedback request/response structures for integrators. - Maintains strict security and routing guidance for agent workflows.
v1.0.5
**Expanded environment/configuration and detailed API guidance added in this release.** - Declared "curl" as a required binary and added optional environment variables (TACE_BASE_URL, TACE_SKILL_VERSION) to metadata. - Replaced general workflow descriptions with explicit API endpoint tables and full authentication storyline. - Documented all key operations: Auth, Catalog, Orders, Payments, Subscriptions, Waitlist, Feedback, and Realtime WebSockets. - Included agent operating loop and precise retry/error-handling procedures. - Clarified catalog rendering rules and product list requirements. - Provided relevant error codes and explicit description of registration, login, and session handling sequence.
v1.0.4
- Simplified and clarified the skill contract for easier human readability. - Limited guidance to high-level workflow categories and core safety rules; removed endpoint specifications and detailed workflow steps. - Made all environment variables optional and removed them from metadata. - Consolidated and shortened safety, routing, and error-handling instructions. - Trimmed individual workflow sections to focus on intent and boundaries, not technical details or deprecated flows.
v1.0.3
**Summary:** Security-focused contract refactor for TACE workflows. - Tightened global data-safety and request boundary rules; added explicit "Never" and "Always" sections. - Clarified routing and contract refresh procedures. - Updated workflow guidance and error handling for auth, discovery, orders, subscriptions, and webhook validation. - Explicitly deprecated merchant onboarding in contract scope (webhook callback only). - Improved product output requirements and idempotency expectations. - Removed shell command execution and dependency on external binaries.
v1.0.2
Version 1.0.2 - Added explicit security posture section, clarifying that this skill is for workflow orchestration only and must never collect, store, or log sensitive authentication data. - Required that all signing for authentication must be performed locally (outside the skill); examples and docs must use masked placeholders for secrets. - Tightened safety rules across authentication flows: secrets, wallet keys, and JWT tokens must never be logged. - Updated auth session workflow and safety guidance to reinforce local signing and proper handling of wallet credentials. - No functional API or routing changes; all updates focus on stricter security and safe secret handling documentation.
v1.0.1
**Big change: All sub-skill specifications now merged into a single unified skill file.** - Consolidated all previous sub-skill files (auth-session, catalog-discovery, orders-payments, subscriptions-waitlist, merchant-onboarding) into one unified `skill.md` contract. - Updated instruction and workflow details to cover all agent flows in a single document. - Revised skill endpoints, global rules, error handling, and data presentation contracts for clarity and consistency. - Deprecated merchant onboarding section—now handled externally, no longer in TACE. - Removed all individual sub-skill files; only the unified contract remains.
v1.0.0
- Initial release of the TACE (Tender Agentic Commerce Engine) top-level skills controller. - Provides task-based routing to sub-skills for auth, catalog, orders, subscriptions, and merchants. - Documents skill endpoints for runtime and static skill contracts. - Specifies global agent rules for versioning, authentication, and error handling. - Outlines expected sub-skill files with their endpoint focuses and responsibilities. - Defines maintenance guidelines to ensure up-to-date skill routing and documentation.
元数据
Slug tace
版本 1.0.6
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 7
常见问题

TACE 是什么?

TACE (Tender Agentic Commerce Engine) runtime contract. Includes concrete payloads required by current endpoints. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 116 次。

如何安装 TACE?

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

TACE 是免费的吗?

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

TACE 支持哪些平台?

TACE 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 TACE?

由 Tender(@tender)开发并维护,当前版本 v1.0.6。

💬 留言讨论