cost-breakdown

1# Cost Breakdown
2 
3Detailed cost analysis from the Agent Monitor's pricing engine.
4 
5## Input
6 
7The user provides: **$ARGUMENTS**
8 
9This may be: "today", "this week", "last 30 days", a session ID, or "budget $50/week".
10 
11## Data Sources
12 
13| Endpoint | Returns |
14|----------|---------|
15| `GET /api/pricing` | `{ pricing: [{ model_pattern, display_name, input_per_mtok, output_per_mtok, cache_read_per_mtok, cache_write_per_mtok }] }` |
16| `GET /api/pricing/cost` | Total cost: `{ total_cost, breakdown: [{ model, input_tokens, output_tokens, cache_read_tokens, cache_write_tokens, cost, matched_rule }] }` |
17| `GET /api/pricing/cost/{sessionId}` | Per-session cost with same breakdown shape |
18| `GET /api/sessions?limit=200` | Sessions list — each includes inline `cost` field (bulk pricing) |
19| `GET /api/analytics` | Token totals (total_input, total_output, total_cache_read, total_cache_write — baselines pre-summed), daily trends |
20 
21### How costs are calculated
22 
23The pricing engine matches model names against `model_pattern` using SQL LIKE (e.g. `claude-sonnet-4-5%` matches `claude-sonnet-4-5-20250514`). **Longest pattern wins** for specificity. Cost per model:
24 
25```
26cost = (input_tokens / 1M) × input_per_mtok
27     + (output_tokens / 1M) × output_per_mtok
28     + (cache_read_tokens / 1M) × cache_read_per_mtok
29     + (cache_write_tokens / 1M) × cache_write_per_mtok
30```
31…