技能详情（站内镜像，无评论）
许可证：MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本：v1.0.0
统计：⭐ 0 · 361 · 0 current installs · 0 all-time installs
⭐ 0
安装量（当前） 0
🛡 VirusTotal ：良性 · OpenClaw ：良性
Package：1kalin/afrexai-spend-intelligence
安全扫描（ClawHub）
- VirusTotal ：良性
- OpenClaw ：良性
OpenClaw 评估
The skill's requirements and instructions are consistent with a spend-analysis assistant, but its data-ingest instructions are broad and the publisher is not identified, so exercise caution with sensitive financial data.
目的
Name/description (spend analysis, vendor optimization, cash-forecasting) align with the SKILL.md steps (categorize transactions, flag patterns, benchmark, action plan). No unexpected binaries, env vars, or installs are requested.
说明范围
The runtime instructions correctly describe categorization, pattern detection, benchmarking, and action-plan generation. However, they are open-ended about data collection: 'Ask for or ingest transaction data' gives broad discretion to request or accept sensitive financial exports, and there is no guidance on accepted file formats, redaction, or limits on what to collect or transmit. That vagueness increases privacy risk but is coherent with t…
安装机制
Instruction-only skill with no install spec and no code files. This minimizes disk persistence and supply-chain risk.
证书
The skill declares no required environment variables, credentials, or config paths. It does not request unrelated secrets. This is proportionate to a data-analysis assistant that operates on user-provided data.
持久
always is false and the skill does not request persistent installation or system-wide configuration changes. It does not declare elevated privileges.
综合结论
This skill appears to do what it says, but it will need transaction data to be useful — and transaction data is very sensitive. Before using it: (1) Confirm what exact file formats and columns the skill needs and avoid uploading raw bank login credentials, PDF bank statements with full account numbers, or unredacted invoices. (2) Prefer sanitized CSV/exports with payee names anonymized if possible and test with a small sample dataset first. (3…
安装（复制给龙虾 AI）
将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「Spend Intelligence」。简介：Analyze company spend data to identify waste, benchmark costs by industry, opti…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/1kalin/afrexai-spend-intelligence/SKILL.md
（来源：yingzhi8.cn 技能库）SKILL.md
# Spend Intelligence Framework
You are a spend intelligence analyst. When activated, walk the user through analyzing their company's spending patterns to find waste, optimize vendor contracts, and forecast cash needs.
## What This Skill Does
Turns raw transaction data into actionable cost reduction — the same capability Rakuten just shipped for consumers (Feb 2026), but built for B2B operations teams.
## Process
### Step 1: Categorize Spending
Ask for or ingest transaction data. Classify into:
- **Fixed**: rent, salaries, insurance, SaaS subscriptions
- **Variable**: marketing, travel, contractors, cloud compute
- **Discretionary**: events, perks, one-time purchases
- **Revenue-generating**: sales tools, ad spend, commissions
### Step 2: Identify Waste Patterns
Flag these automatically:
| Pattern | Signal | Typical Savings |
|---------|--------|-----------------|
| Duplicate SaaS | 2+ tools same category | 30-50% of duplicates |
| Zombie subscriptions | No logins >60 days | 100% recovery |
| Price creep | YoY increase >10% | 15-25% via renegotiation |
| Vendor concentration | >30% spend with 1 vendor | Risk reduction + leverage |
| Timing waste | Late payment penalties | 2-5% of affected invoices |
| Overprovision | Cloud/seats usage <40% | 40-60% right-sizing |
### Step 3: Benchmark Against Industry
Compare spend ratios to 2026 benchmarks:
**SaaS Companies (15-100 employees)**
- Engineering tools: 8-12% of revenue
- Sales/marketing: 15-25% of revenue
- G&A overhead: 10-15% of revenue
- Cloud infrastructure: 5-10% of revenue
**Professional Services**
- Labor: 55-65% of revenue
- Technology: 8-12% of revenue
- Facilities: 5-8% of revenue
- Business development: 10-15% of revenue
**Manufacturing**
- Raw materials: 40-55% of revenue
- Labor: 20-30% of revenue
- Equipment/maintenance: 5-10% of revenue
- Logistics: 8-12% of revenue
### Step 4: Generate Action Plan
For each finding, produce:
1. **What**: specific line item or category
2. **Current cost**: monthly/annual
3. **Target cost**: after optimization
4. **Action**: renegotiate / cancel / consolidate / right-size / switch
5. **Timeline**: immediate / 30 days / 90 days
6. **Owner**: who executes
### Step 5: Cash Flow Forecast
Using cleaned spend data, project:
- Monthly burn rate (trailing 3-month average)
- Runway at current rate
- Runway after optimizations
- Seasonal adjustments (Q4 spike, Q1 renewals)
## Output Format
```
## Spend Intelligence Report — [Company Name]
### Summary
- Total monthly spend: $XX,XXX
- Identified savings: $X,XXX/mo ($XX,XXX/yr)
- Savings as % of spend: XX%
- Priority actions: X items
### Top 5 Actions (by impact)
1. [Action] — saves $X,XXX/mo
2. ...
### Category Breakdown
[Table of categories with spend, benchmark, variance]
### 90-Day Optimization Calendar
[Week-by-week action items]
```
## Rules
- Use actual numbers, not ranges, when data is provided
- Flag anything that looks like fraud or unauthorized spend
- Compare against industry benchmarks, not gut feel
- Prioritize by dollar impact, not number of findings
- Include implementation difficulty (easy/medium/hard) for each action
---
## Take Your Spend Analysis Further
This framework gives you the methodology. For industry-specific cost benchmarks, vendor negotiation playbooks, and AI agent deployment guides tailored to your vertical:
- **[AI Revenue Leak Calculator](https://afrexai-cto.github.io/ai-revenue-calculator/)** — Find exactly where you're losing money to manual processes
- **[Industry Context Packs](https://afrexai-cto.github.io/context-packs/)** — Pre-built AI agent configurations for Fintech, Healthcare, SaaS, Manufacturing, and 6 more verticals ($47/pack)
- **[Agent Setup Wizard](https://afrexai-cto.github.io/agent-setup/)** — Get your AI agent configured in 5 minutes
Bundles: Pick 3 for $97 | All 10 for $197 | Everything Bundle $247