How do I install the Cursor Usage skill?

Run npx skillmds@latest add nirjhar17/cursor-usage in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Cursor Usage safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Cursor Usage?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Cursor Usage free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under MIT.

Who published Cursor Usage?

nirjhar17 (@nirjhar17) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Cursor Usage by nirjhar17 — Agent Skill

Name: Cursor Usage
Author: nirjhar17

Cursor Usage

Show Cursor agent token usage, estimated costs, and activity patterns across all sessions on this machine. Supports summary, by-project, by-month, by-session, daily-activity, and tool-usage views. Use when asked about spending, costs, tokens, daily activity, usage patterns, or missing days.

nirjhar17 Updated Sep 21, 2026

File contents

Cursor Agent Usage & Cost Report

Show estimated token usage and costs across all Cursor agent sessions stored on this machine (~/.cursor/projects/*/agent-transcripts/*/*.jsonl).

How It Works

Cursor agent transcripts don't include raw token counts like Claude Code sessions. This skill estimates tokens from character length (~4 characters per token) and applies Anthropic's published Sonnet pricing to approximate costs. It also tracks message counts, tool usage, and activity patterns.

Trigger Prompts

Run this skill when the user asks about any of these topics:

"How much have I spent on Cursor?"
"Show my Cursor usage / costs / tokens"
"What's my daily activity?"
"How much did I spend this week / month?"
"Which projects cost the most?"
"Show my most expensive sessions"
"Any days I didn't use Cursor?"
"What are my usage patterns?"
"What tools do I use the most?"
"/usage"

Reports Available

Based on what the user asks, run the appropriate script from this skill's scripts/ directory:

Summary (default)

python3 scripts/tally-costs.py

Shows total sessions, estimated tokens, and estimated cost. Use when the user asks a general "how much" question.

By Project

python3 scripts/tally-costs.py --by-project

Use when the user asks which projects cost the most or where tokens are being spent.

By Month

python3 scripts/tally-costs.py --by-month

Use when the user asks about monthly trends or spending over time.

By Session (Top 20)

python3 scripts/tally-costs.py --by-session

Use when the user asks about their most expensive sessions.

Tool Usage

python3 scripts/tally-costs.py --tools

Use when the user asks which tools are used most, or wants a breakdown of Shell vs Read vs Write etc.

Daily Activity

python3 scripts/daily-activity.py

Use when the user asks about daily patterns, missing days, heavy usage days, or day-of-week breakdown.

Presenting Results

Run the appropriate script and show the output directly — it's already formatted for terminal display.
If the user asks a specific question (e.g., "how much did I spend last week?"), extract the relevant data from the output and answer concisely rather than dumping the full table.
If the user asks to compare projects or time periods, run multiple reports if needed.
Remind the user these are estimates based on character-to-token approximation and published per-token pricing — actual billing will differ.

Limitations

Token counts are approximated (~4 chars/token) since Cursor transcripts don't include API usage metadata
Does not account for system prompts, context window overhead, or cache read/write pricing
Only captures sessions with agent transcripts (excludes inline completions, tab completions, etc.)
Cost estimates assume Sonnet pricing; actual model mix may vary
Session data availability depends on Cursor's transcript retention policy

---name: cursor-usagedescription: Show Cursor agent token usage, estimated costs, and activity patterns across all sessions on this machine. Supports summary, by-project, by-month, by-session, daily-activity, and tool-usage views. Use when asked about spending, costs, tokens, daily activity, usage patterns, or missing days.license: MIT---# Cursor Agent Usage & Cost ReportShow estimated token usage and costs across all Cursor agent sessions stored on this machine (`~/.cursor/projects/*/agent-transcripts/*/*.jsonl`).## How It WorksCursor agent transcripts don't include raw token counts like Claude Code sessions. This skill estimates tokens from character length (~4 characters per token) and applies Anthropic's published Sonnet pricing to approximate costs. It also tracks message counts, tool usage, and activity patterns.## Trigger PromptsRun this skill when the user asks about any of these topics:- "How much have I spent on Cursor?"- "Show my Cursor usage / costs / tokens"- "What's my daily activity?"- "How much did I spend this week / month?"- "Which projects cost the most?"- "Show my most expensive sessions"- "Any days I didn't use Cursor?"- "What are my usage patterns?"- "What tools do I use the most?"- "/usage"## Reports AvailableBased on what the user asks, run the appropriate script from this skill's `scripts/` directory:### Summary (default)```bashpython3 scripts/tally-costs.py```Shows total sessions, estimated tokens, and estimated cost. Use when the user asks a general "how much" question.### By Project```bashpython3 scripts/tally-costs.py --by-project```Use when the user asks which projects cost the most or where tokens are being spent.### By Month```bashpython3 scripts/tally-costs.py --by-month```Use when the user asks about monthly trends or spending over time.### By Session (Top 20)```bashpython3 scripts/tally-costs.py --by-session```Use when the user asks about their most expensive sessions.### Tool Usage```bashpython3 scripts/tally-costs.py --tools```Use when the user asks which tools are used most, or wants a breakdown of Shell vs Read vs Write etc.### Daily Activity```bashpython3 scripts/daily-activity.py```Use when the user asks about daily patterns, missing days, heavy usage days, or day-of-week breakdown.## Presenting Results- Run the appropriate script and show the output directly — it's already formatted for terminal display.- If the user asks a specific question (e.g., "how much did I spend last week?"), extract the relevant data from the output and answer concisely rather than dumping the full table.- If the user asks to compare projects or time periods, run multiple reports if needed.- Remind the user these are **estimates** based on character-to-token approximation and published per-token pricing — actual billing will differ.## Limitations- Token counts are approximated (~4 chars/token) since Cursor transcripts don't include API usage metadata- Does not account for system prompts, context window overhead, or cache read/write pricing- Only captures sessions with agent transcripts (excludes inline completions, tab completions, etc.)- Cost estimates assume Sonnet pricing; actual model mix may vary- Session data availability depends on Cursor's transcript retention policy

Frequently asked questions