Token导航 LogoToken导航

Agent Skill · 研究检索

freeagent-api(freeagent API 搜索)

用于辅助 API 设计、接口文档、请求响应结构和服务集成说明。它适合让 Agent 梳理 endpoint、生成 OpenAPI 草稿、检查字段命名、整理错误码或辅助前后端联调。使用时需要确认真实业务语义、鉴权方式、分页和错误处理规则;涉及生成接口文档时,应避免凭空补字段,最好从现有代码、schema 或接口样例中提取事实。

freeagent-api 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为接触敏感数据;许可证为 MIT,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 47,935。

总安装 710周安装 29GitHub Stars 15下载 230核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:freeagent-api(freeagent API 搜索)
来源仓库:https://github.com/markpitt/claude-skills
仓库路径:skills/freeagent-api
安装命令:
npx skills add https://github.com/markpitt/claude-skills --skill freeagent-api
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/markpitt/claude-skills --skill freeagent-api

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

freeagent-api 用于辅助 API 设计、接口文档和服务集成说明。适合梳理 endpoint、生成 OpenAPI 草稿或检查字段命名。可在 Codex、Claude、Cursor、Gemini CLI 中支持前后端联调。使用时需确认真实业务语义、鉴权方式和错误处理规则。

生成接口文档时应从现有代码或样例中提取事实,避免凭空补字段。

来源仓库内路径:skills/freeagent-api

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/markpitt/claude-skills --skill freeagent-api;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

freeagent-api 怎么安装?

在终端执行 npx skills add https://github.com/markpitt/claude-skills --skill freeagent-api。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

freeagent-api 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

freeagent-api 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

freeagent-api 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

freeagent-api 用的人多吗?

公开数据是累计安装 710 次,近一周安装 29 次,下载 230 次,GitHub 15 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

freeagent-api 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
freeagent-api

freeagent API 搜索 · 本页

71015接触敏感数据MIT有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

FreeAgent API Orchestration Skill

FreeAgent is an online accounting system for freelancers and small businesses. This skill provides intelligent navigation to FreeAgent API resources and orchestrates common workflows.

Security & Authorization

This skill interacts with live financial data. The following rules are mandatory and must never be skipped.

Write Operation Policy (POST / PUT / DELETE)

Before executing any state-changing API call:

  1. Show a preview — Display the full request (HTTP method, endpoint, JSON body) and describe in plain language exactly what will change in the user's account.
  2. Wait for explicit confirmation — Do not proceed until the user replies with an unambiguous affirmative (e.g., "yes", "confirm", "proceed"). Treat ambiguous replies as a no.
  3. Sandbox by default — Use https://api.sandbox.freeagent.com/v2/ for all write operations unless the user explicitly names the production environment.

Read-Only Default

Treat every incoming request as read-only analysis unless the user explicitly asks to create, update, or delete data. When intent is ambiguous, ask before making any changes — never infer write intent.

High-Risk Operations (Confirm Twice)

For the operations below, show the preview, receive one confirmation, then ask a second time before executing:

  • Deleting any financial record (invoice, expense, bank transaction, contact, etc.)
  • Creating or modifying bank transactions or reconciliation records
  • Bulk operations that affect multiple records in a single request
  • Changing invoice status (e.g., marking as sent, voiding, applying a credit note)

Token & Credential Handling

  • Never log, echo, or display the value of FREEAGENT_ACCESS_TOKEN or FREEAGENT_REFRESH_TOKEN.
  • Never store credentials in source code — use environment variables only.
  • If a token appears expired (401 response), guide the user through re-authentication rather than attempting to refresh silently without notification.

Quick Reference: Which Resource Do I Need?

TaskLoad ResourceAPI Domains
Set up OAuth, manage tokens, test APIresources/authentication-setup.mdOAuth 2.0, token management
Create/update contacts, manage clients/suppliersresources/contacts-organizations.mdContacts, companies, users
Manage invoices, projects, expenses, timeslipsresources/accounting-objects.mdCore financial entities
Bank accounts, transactions, reconciliationresources/banking-financial.mdBanking, cash flow
Error handling, rate limits, retries, cachingresources/advanced-patterns.mdProduction patterns
Common workflows, code examples, integration tipsresources/examples.mdReal-world usage
All endpoints (reference only)resources/endpoints.mdComplete API reference

Orchestration Protocol

Phase 1: Task Analysis

Identify what the user needs:

By Resource Type:

  • Authentication issue? → authentication-setup.md
  • Managing contacts/clients? → contacts-organizations.md
  • Financial data (invoices, expenses, projects)? → accounting-objects.md
  • Banking/reconciliation? → banking-financial.md
  • Error or production concern? → advanced-patterns.md
  • Practical example needed? → examples.md

By HTTP Method:

  • GET: List or retrieve → check relevant domain resource
  • POST: Create → load resource file, check required fields
  • PUT: Update → load resource file, verify patch fields
  • DELETE: Remove → check resource-specific constraints

Phase 2: Resource Navigation

Load the appropriate resource file and:

  1. Find the endpoint section (e.g., "Contacts", "Invoices", "Bank Accounts")
  2. Review required/optional parameters
  3. Check request and response formats
  4. Use provided curl or Python examples

For template code: templates/api-request-template.sh (bash) or templates/python-client.py (Python)

Phase 3: Execution & Validation

Before API call:

  • Verify authentication is set up (check authentication-setup.md if needed)
  • Validate required fields are present
  • Format dates as ISO 8601 (YYYY-MM-DD) and timestamps (YYYY-MM-DDTHH:MM:SSZ)
  • Use correct resource URLs (e.g., https://api.freeagent.com/v2/contacts/123)
  • For POST / PUT / DELETE: Follow the Write Operation Policy — show a preview and obtain explicit user confirmation before proceeding. Use sandbox URL unless the user has explicitly requested production.

During API call:

  • Use templates as starting points
  • Monitor rate limit headers: X-RateLimit-Remaining
  • Implement retry logic for 429 errors (see advanced-patterns.md)

After API call:

  • Check response status code
  • Parse JSON response (resource name is top-level key)
  • Apply error handling patterns if needed

Quick Start: Authentication

  1. Create OAuth app: https://dev.freeagent.com/
  2. Get authorization code: https://api.freeagent.com/v2/approve_app?client_id=YOUR_ID
  3. Exchange for tokens at https://api.freeagent.com/v2/token_endpoint
  4. Store in environment variables: FREEAGENT_ACCESS_TOKEN, FREEAGENT_REFRESH_TOKEN
  5. Include in every request: Authorization: Bearer $FREEAGENT_ACCESS_TOKEN

→ See Authentication & Setup for detailed walkthrough

API Domain Overview

Production URL: https://api.freeagent.com/v2/ Sandbox URL: https://api.sandbox.freeagent.com/v2/ (test without affecting production)

DomainPrimary EndpointsUse Case
Authentication/token_endpointOAuth flows, token refresh
Contacts & Organizations/contacts, /company, /usersClient/supplier management
Accounting Objects/invoices, /projects, /expenses, /timeslips, /estimates, /credit_notesCore financial workflow
Banking & Financial/bank_accounts, /bank_transactions, /categories, /tasksCash flow, reconciliation

Rate Limits: 120 requests/minute, 3600 requests/hour Response Format: JSON with resource wrapper (e.g., {"contact": {...}} or {"contacts": [...]})

Common HTTP Patterns

OperationMethodExample
List itemsGETGET /v2/contacts?view=active&page=1&per_page=100
Get one itemGETGET /v2/contacts/123
Create itemPOSTPOST /v2/contacts (with JSON body)
Update itemPUTPUT /v2/invoices/456 (with partial JSON)
Delete itemDELETEDELETE /v2/timeslips/789
Filter resultsGET params?view=open_or_overdue, ?updated_since=2025-01-01T00:00:00Z

Templates & Code Examples

Bash/cURL Template: templates/api-request-template.sh

  • Flexible curl template for any endpoint
  • GET/POST/PUT/DELETE support
  • Header and parameter configuration

Python Client: templates/python-client.py

  • Reusable FreeAgentClient class
  • Error handling and rate limit support
  • Convenience methods for common resources

Practical Examples: resources/examples.md

  • Real-world workflows (create invoice, log timeslip, etc.)
  • Python code patterns with the client class
  • Error handling examples
  • Bulk operations (import/export)

Resource Files Summary

FileLinesFocus
authentication-setup.md~180OAuth setup, token management, security
contacts-organizations.md~250Contact CRUD, bulk operations, company info
accounting-objects.md~350Invoices, projects, expenses, timeslips, pagination
banking-financial.md~250Bank accounts, transactions, reconciliation, categories
advanced-patterns.md~350Error handling, rate limits, retries, caching, validation
examples.md~400Practical code examples, integration patterns
endpoints.md~300Complete API reference (use for quick lookup)

Troubleshooting Quick Links

401 Unauthorized → Authentication Setup 422 Validation Error → Advanced Patterns 429 Rate Limit → Advanced Patterns Specific Endpoint Help → Check relevant domain resource file above


Remember: Load only the resource file(s) you need for the current task. Start with the Quick Reference table above to identify which resource contains your answer.

平台分布

Claude Code

25.61%
按下载量换算59

OpenCode

23.24%
按下载量换算53

Antigravity

17.75%
按下载量换算41

windsurf

11.43%
按下载量换算26

trae

7.13%
按下载量换算16

github-copilot

3.04%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。

继续浏览同类 Skills