← 返回 Skills 市场
327
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install aerobase-awards
功能描述
Search 24+ airline loyalty programs for award space with miles cost, seat availability, and jetlag scores
使用说明 (SKILL.md)
Award Flight Intelligence
Search
POST /api/v1/awards/search — search seats.aero cache (286K+ rows, 24 programs)
Body: { origin?, destination?, program?, cabin?, dateFrom?, dateTo?, minScore? }
Returns: jetlagScore (0-10 raw, multiply by 10 for K2), k2CabinScore, k2BaseScore,
milesCost, remaining_seats, transfer_options.
Cabin values: economy, business, premium, first (full names, NOT codes).
Saved Award Trips
GET /api/v1/awards/trips — retrieve saved award trip search results Returns previously searched award itineraries with segments, pricing, and jetlag scores.
Alerts
- POST /api/awards/alerts — create alert
- GET /api/awards/alerts — list user's alerts
- Continuous alerts require Pro tier
Program Intelligence
24 programs: Copa, Aeromexico, Ethiopian, United, Delta, American, BA, Lufthansa, AF/KLM, etc. Cross-reference user's credit card portfolio with transfer partner map.
Always
- Present prices in miles/points, not cash
- Include jetlag score alongside every result
- Warn about fuel surcharges on BA, LH, AF awards
- Calculate cents-per-point value: "Business to Tokyo is 75K Aeroplan. Cash price is $3,200. That's 4.3 cents/point — excellent value, your points are worth more here than average."
- When comparing cash vs. points: "The economy option saves 50K miles but the 11 PM departure scores 7.8 — four days of zombie mode. The business seat literally buys you three productive days."
Rate Limits
- Award search: max 20/hr. Alert create: max 10 alerts per user total.
- Award monitoring cron: every 4 hours (do not increase). Batch route queries where possible.
Scrapling — Cash Price Lookup for CPP Calculation
Use Scrapling /search to get cash prices from Google Flights for cents-per-point value:
Reference: Scrapling Documentation
POST {SCRAPLING_URL}/search
{"site":"google-flights","origin":"JFK","destination":"LHR","departure":"2026-04-01","return":"2026-04-08"}
Returns: {"results": [{"price":"€488","duration":"6 hr 50 min","stops":"Nonstop"}], "count": 48}
CPP Calculation Workflow
- Get award price from seats.aero API (e.g., 60K Aeroplan)
- Fire Scrapling Google Flights search for same route
- Extract cash price (EUR from Helsinki — convert to USD if needed)
- Calculate:
cash_price_usd / miles_required * 100 = cents_per_point - Present: "Business to London is 60K Aeroplan. Cash price is $543. That's 0.9 cpp."
seats.aero (API preferred, browser fallback with PROXY)
Cloudflare protected. Use API cache first. Browser via proxy only for logged-in users.
Airline Award Search (PROXY required)
- Navigate to airline's FFP booking page (behind login wall)
- User must provide credentials for their account
- NEVER store or log FFP credentials
When to SKIP browser entirely
- General award search → seats.aero API cache is comprehensive
- Award alerts → API handles this automatically
- Program comparisons → API data is sufficient
- Only use browser for visual verification or cash-price lookup
安全使用建议
This skill broadly does what it says (award searches), but its instructions ask for sensitive data and reference undeclared endpoints/config. Before installing or using it, consider: 1) The skill asks you to provide airline FFP credentials and to supply (or let it read) your credit-card portfolio — sharing passwords or card data is risky; prefer OAuth or token-based access from the airline if available. 2) The SKILL.md calls for a Scrapling endpoint ({SCRAPLING_URL}) and a proxy for browser access but those configuration variables are not declared — ask the author what exact environment variables or URLs the skill needs and how they are secured. 3) The skill says 'NEVER store or log FFP credentials' — that is a policy, not an enforcement; confirm how the agent handles secrets at runtime and whether input fields are ephemeral and secured. 4) The source/homepage is unknown; request the source code or a trusted homepage and an explanation of how browser proxying is implemented and who operates the proxy/Scrapling endpoint. 5) If you must use it, avoid typing primary account passwords directly; prefer creating limited-access or view-only accounts, use per-service app passwords if available, or use official APIs. If you cannot validate the external endpoints and the author, treat this as high-risk and do not provide real credentials or financial data.
功能分析
Type: OpenClaw Skill
Name: aerobase-awards
Version: 1.0.0
The skill bundle instructs the AI agent to handle sensitive airline Frequent Flyer Program (FFP) credentials to access booking pages behind login walls in SKILL.md. While the instructions explicitly state to 'NEVER store or log FFP credentials,' the design pattern of passing plaintext credentials through an LLM-driven agent to third-party sites is inherently high-risk. Additionally, the skill utilizes an external scraping service via a variable '{SCRAPLING_URL}' and requires browser automation with proxies, which significantly expands the attack surface for credential interception.
能力评估
Purpose & Capability
The stated purpose (search award inventory across programs, show miles cost and jetlag scores) matches the instructions in general. However, the SKILL.md also calls for cross-referencing a user's credit-card portfolio and using browser+proxy access to airline FFP pages; those extra capabilities are plausible add-ons but are not reflected in the declared requirements (only AEROBASE_API_KEY is declared).
Instruction Scope
The SKILL.md instructs the agent to: call a Scrapling service at {SCRAPLING_URL}, use a browser via a proxy for logged-in pages, and request user FFP credentials (with an instruction to not store them). It also says to cross-reference the user's credit-card portfolio. These are sensitive actions and the SKILL.md references env/config values (SCRAPLING_URL, PROXY settings, where/how to supply user FFP credentials and card portfolio) that are not declared, leaving open how the agent will obtain and protect that data.
Install Mechanism
Instruction-only skill with no install spec or code files; nothing is written to disk by the skill itself. This is lower-risk from an installation perspective.
Credentials
Only AEROBASE_API_KEY is declared as primaryEnv, but the runtime instructions clearly rely on additional external endpoints/config (SCRAPLING_URL, PROXY) and require per-user sensitive inputs (airline login credentials, credit-card portfolio). The undeclared/unspecified inputs are disproportionate to the declared environment surface.
Persistence & Privilege
The skill is not marked always:true and has no install modifications. Autonomous invocation is allowed by platform default — combine that fact with the sensitive data flows above before deciding to enable autonomous use.
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 aerobase-awards - 安装完成后，直接呼叫该 Skill 的名称或使用
/aerobase-awards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
v1.0.0
- Initial release of aerobase-awards skill.
- Search over 24 major airline loyalty programs for award seats with miles cost, seat availability, and jetlag scores.
- View saved award trip searches and set up personalized award alerts (requires Pro tier for continuous monitoring).
- See price breakdowns in miles/points, automatic jetlag scoring, warning for fuel surcharges, and cents-per-point (cpp) value calculations using real cash prices from Google Flights via Scrapling.
- User rate limits: max 20 searches/hr, 10 alerts/user; award monitoring every 4 hours.
- Secure workflow for airline website access; credentials never stored or logged.
元数据
常见问题
Aerobase Awards 是什么？
Search 24+ airline loyalty programs for award space with miles cost, seat availability, and jetlag scores. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 327 次。
如何安装 Aerobase Awards？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install aerobase-awards」即可一键安装，无需额外配置。
Aerobase Awards 是免费的吗？
是的，Aerobase Awards 完全免费（开源免费），可自由下载、安装和使用。
Aerobase Awards 支持哪些平台？
Aerobase Awards 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 Aerobase Awards？
由 Aerobase（@kurosh87）开发并维护，当前版本 v1.0.0。
推荐 Skills