LLM Cost Tracker
Per-group, per-user, and per-model cost attribution for your LLM API spend. No proxy. No infrastructure. SDK-level enforcement.
https://llmcosttracker.comOpens ChatGPT on the web or desktop and asks it to use the WebMCP tools available here.
Connect straight to this server’s public endpoint.
https://llmcosttracker.com/api/mcpWe add this server to your workspace, then open Studio — saved access, one connection to many servers, with a history of what ran.
Last probed Sep 14, 2026 · llmcosttracker.com
7tools discovered
Lct Get Overview
Get top-level LLM cost KPIs for a project: total spend, call count, average cost per call, and average tokens per call. Includes period-over-period deltas. Use this as the starting point to understand overall spend before drilling into features or users. Args: - api_key: LLM Cost Tracker project API key (starts with lct_live_) - range: Time window — 'today', '7d', '30d', or 'all' (default: '30d') Returns: total_cost_usd, total_calls, avg_cost_per_call, avg_tokens_per_call, delta (period-o
Lct Get Features
Get LLM cost attribution grouped by feature tag (e.g. 'search', 'chat', 'summarize'). Sorted by total spend descending. Use this to understand which product features are driving LLM costs. Args: - api_key: LLM Cost Tracker project API key - range: Time window — 'today', '7d', '30d', or 'all' (default: '30d') Returns array of features with: name, total_cost_usd, call_count, avg_cost_per_call, avg_tokens, avg_latency_ms, pct_of_total
Lct Get Feature Detail
Get a full cost breakdown for a single feature tag, including model breakdown, top users, prompt versions, and daily spend timeseries. Use this to drill into a specific feature after identifying it as expensive via lct_get_features. Args: - api_key: LLM Cost Tracker project API key - feature: Feature tag to retrieve (e.g. 'search', 'chat') - range: Time window — 'today', '7d', '30d', or 'all' (default: '30d') Returns: KPIs, delta_pct, models breakdown, top_users (top 5), versions compar
Lct Get Users
Get LLM cost attribution grouped by user ID, sorted by total spend. Includes anomaly signals for users whose cost-per-call is significantly above or below the fleet average. Args: - api_key: LLM Cost Tracker project API key - range: Time window — 'today', '7d', '30d', or 'all' (default: '30d') - limit: Max users to return, 1-1000 (default: 50) Returns: data array with id, total_cost_usd, call_count, avg_cost_per_call, pct_of_total, signal ('expensive'|'cheap'|null), signal_multiple. Plus
Lct Get User Detail
Get a full cost and usage breakdown for a single user ID, including wasted spend, feature breakdown, and daily timeseries. Use this to investigate a specific user after identifying them as high-spend via lct_get_users. Args: - api_key: LLM Cost Tracker project API key - user_id: User ID to retrieve. Use 'anonymous' for untagged calls. - range: Time window — 'today', '7d', '30d', or 'all' (default: '30d') Returns: KPIs, is_high_spend, wasted_cost_usd, wasted_calls, first_seen, last_seen,
Lct Get Models
Get LLM cost attribution grouped by model, sorted by total spend. Includes provider information. Args: - api_key: LLM Cost Tracker project API key - range: Time window — 'today', '7d', '30d', or 'all' (default: '30d') Returns array of models with: model, provider, total_cost_usd, call_count, avg_cost_per_call, pct_of_total
Lct Get Timeseries
Get daily LLM spend totals over a time period, optionally filtered to a single feature. Each bucket is one UTC calendar day. Args: - api_key: LLM Cost Tracker project API key - range: Time window — 'today', '7d', '30d', or 'all' (default: '30d') - feature: Optional feature tag to filter to (e.g. 'search') Returns array of daily buckets with: date (YYYY-MM-DD), total_cost_usd, call_count
Get your MCP into directories
A working endpoint is step one. Directory coverage is the coordinated launch across ChatGPT, Claude, Cursor, the MCP Registry, and community indexes.
Directory coverage for brandsPer-group, per-user, and per-model cost attribution for your LLM API spend. No proxy. No infrastructure. SDK-level enforcement.
Use the MCP endpoint listed on this page in your MCP client configuration. One-click install pills support Claude, Cursor, VS Code, and other hosts. Copy the remote MCP URL if your client needs a manual entry.
MCPBundles probed 7 tools on the live server. The tool list on this page reflects what was discovered at the last refresh — connect your client to see the full set available to your session.
No provider sign-in was required during MCPBundles' probe. Your client may still need MCPBundles credentials depending on how you connect.
Operate LLM Cost Tracker? Verify ownership to take over this directory entry.
This server appears in the MCPBundles directory. Verify you operate it to take over the listing — name, description, logo, contact email, and skill content. We email a 6-digit code to a maintainer address your server publishes in /.well-known/security.txt or /.well-known/mcpbundles.json. Free, takes about a minute.