← 返回 Skills 市场
dwflickinger

AgentClear

作者 dwflickinger · GitHub ↗ · v1.1.0
cross-platform ✓ 安全检测通过
266
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install agentclear
功能描述
Discover, call, and pay for APIs through AgentClear — the commerce layer for AI agents. Use when you need to: (1) find an API or tool by describing what you...
使用说明 (SKILL.md)

AgentClear — Agent Commerce Layer

Discover and call paid APIs with natural language. One API key, per-call billing, zero friction.

Setup

Requires AGENTCLEAR_API_KEY environment variable (starts with axk_).

export AGENTCLEAR_API_KEY="axk_your_key_here"

Get a free key with $5 credits: https://agentclear.dev/login

Data & Privacy

  • Discovery queries are used for semantic matching and logged to power the Bounty Board (demand signals for missing services). No PII is collected.
  • Proxy calls forward your payload to the upstream service provider you selected. AgentClear does not store request/response payloads — only billing metadata (service ID, timestamp, cost).
  • API keys authenticate and meter usage. Keys are scoped to your account and can be revoked at any time.
  • All traffic is over HTTPS. See https://agentclear.dev/security for full details.

Endpoints

Base URL: https://agentclear.dev

Discover Services

Find APIs by describing what you need:

curl -X POST https://agentclear.dev/api/discover \
  -H "Authorization: Bearer $AGENTCLEAR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"query": "parse invoices from PDF", "limit": 5}'

Response returns ranked services with id, name, description, price_per_call, and trust_score.

Call a Service

Proxy a request through AgentClear (auto-billed per call):

curl -X POST https://agentclear.dev/api/proxy/{service_id} \
  -H "Authorization: Bearer $AGENTCLEAR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"your": "payload"}'

The payload is forwarded to the upstream service. Response includes the service result plus billing metadata.

List Services

Browse available services:

curl https://agentclear.dev/api/services \
  -H "Authorization: Bearer $AGENTCLEAR_API_KEY"

Workflow

  1. Discover — Describe what you need → get ranked service matches
  2. Evaluate — Check price, trust score, and description
  3. Call — Proxy the request → get result + pay per call
  4. If no service exists, check the Bounty Board at https://agentclear.dev/bounties

Pricing

  • Services range from $0.001 to $1+ per call (provider-set)
  • Platform fee: 2.5%
  • New accounts get $5 free credits
  • Balance auto-deducts per call

Tips

  • Use specific, descriptive queries for better discovery results ("parse QuickBooks QBO files to JSON" beats "parse files")
  • Chain services: discover → call → use result as input to another service
  • If a discover query returns no results, the query is logged as demand on the Bounty Board — providers will build it
安全使用建议
This skill appears coherent for calling the AgentClear marketplace, but be aware: (1) each proxy call may bill your account per-call — ensure you understand pricing and set sensible limits, (2) payloads you send are forwarded to upstream providers, so do not include sensitive secrets or personal data unless you trust the target and AgentClear's policies, (3) SKILL.md asserts discovery queries contain no PII and proxy payloads are not stored — you should verify the vendor's privacy/security docs and consider using a throwaway/test account or revocable API key while evaluating, and (4) when installing from the README's GitHub links, prefer to inspect the fetched SKILL.md content locally before enabling it in your agent.
功能分析
Type: OpenClaw Skill Name: agentclear Version: 1.1.0 The AgentClear skill is a legitimate commerce layer for AI agents to discover and interact with paid APIs. It uses a dedicated API key (AGENTCLEAR_API_KEY) to facilitate semantic discovery and proxied API calls through agentclear.dev, with no evidence of data exfiltration, malicious execution, or harmful prompt injection.
能力评估
Purpose & Capability
Name/description match the declared requirements: the skill is an instruction-only wrapper for the AgentClear API and only requires AGENTCLEAR_API_KEY to operate. No unrelated env vars, binaries, or install steps are requested.
Instruction Scope
SKILL.md instructs the agent to send discovery queries and arbitrary request payloads to AgentClear's proxy endpoints (https://agentclear.dev). This is expected for a proxy/marketplace skill, but forwarding arbitrary payloads to upstream services can expose sensitive data if the agent includes it in requests; SKILL.md claims discovery queries contain no PII and proxy calls are not stored, which the user must trust the vendor for.
Install Mechanism
No install spec or code is included (instruction-only), so nothing is written to disk by an installer. README suggests cloning a GitHub repo or downloading SKILL.md from raw.githubusercontent.com which is common and not unusual—but fetching files from remote sources always carries the usual supply-chain caution.
Credentials
Only a single AGENTCLEAR_API_KEY is required and that directly maps to the described functionality (authorizing calls to AgentClear). No extra secrets or unrelated credentials are requested.
Persistence & Privilege
always is false and the skill is user-invocable with normal autonomous invocation allowed. The skill does not request permanent system-wide privileges or modifications to other skills.
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 agentclear
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /agentclear 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
v1.1.0
Added required env credential declaration (AGENTCLEAR_API_KEY) in metadata. Added Data & Privacy section with transparency on query logging, payload handling, and key scoping.
v1.0.0
Initial release: discover, call, and pay for APIs with natural language
元数据
Slug agentclear
版本 1.1.0
许可证
累计安装 0
当前安装数 0
历史版本数 2
常见问题

AgentClear 是什么?

Discover, call, and pay for APIs through AgentClear — the commerce layer for AI agents. Use when you need to: (1) find an API or tool by describing what you... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 266 次。

如何安装 AgentClear?

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

AgentClear 是免费的吗?

是的,AgentClear 完全免费(开源免费),可自由下载、安装和使用。

AgentClear 支持哪些平台?

AgentClear 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 AgentClear?

由 dwflickinger(@dwflickinger)开发并维护,当前版本 v1.1.0。

💬 留言讨论