enterprise-rag-knowledge-systems
Skill4 starsChanged 3 months agoMITView on GitHub
Real-time COGS metering (calculating exact $ cost per voice minute, SMS unit, LLM token, and per-tenant margin analytics) for multi-tenant AI SaaS platforms.
In this repository19 other files it gives its agents.
- agent-introspection-debuggingSkill
- airops-programmatic-content-engineSkill
- api-designSkill
- code-review-and-qualitySkill
- eval-harnessSkill
- geo-aeo-agent-optimizationSkill
- langchain-agent-patternsSkill
- mcp-server-patternsSkill
- multi-agent-crew-patternsSkill
- openai-agents-sdkSkill
- pgvector-hybrid-searchSkill
- prd-criticSkill
- product-spec-generatorSkill
- prompt-engineerSkill
- security-and-hardeningSkill
- security-reviewSkill
- signal-based-ai-agentSkill
- tdd-workflowSkill
- ui-ux-pro-maxSkill
The file
Files
📊 COGS Metering & Usage Billing Skill
This skill defines the telemetry and real-time cost metering architecture for multi-tenant AI SaaS platforms.
🛠️ COGS Calculation Blueprint
For every voice call, SMS conversation, or AI workflow, compute exact operational cost (COGS):
$$\text{COGS} = (\text{Voice Mins} \times C_{\text{voice}}) + (\text{SMS Units} \times C_{\text{sms}}) + (\text{Tokens}{\text{in}} \times C{\text{in}}) + (\text{Tokens}{\text{out}} \times C{\text{out}})$$
Where default cost rates are:
- $C_{\text{voice}} = $0.07 / \text{min}$ (Retell AI / Telephony)
- $C_{\text{sms}} = $0.0079 / \text{msg}$ (Twilio / Telnyx)
- $C_{\text{in}} = $0.00000025 / \text{token}$ (Claude 3.5 Haiku Input)
- $C_{\text{out}} = $0.00000125 / \text{token}$ (Claude 3.5 Haiku Output)
💻 Sample Metering Record (JSON)
{
"tenant_id": "rooftop_881",
"call_id": "call_abc992",
"duration_seconds": 184,
"usage": {
"voice_minutes": 3.06,
"voice_cost_usd": 0.214,
"llm_tokens_total": 3450,
"llm_cost_usd": 0.0042,
"telephony_cost_usd": 0.045
},
"total_cogs_usd": 0.2632
}Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

