skills.rest
crisandrews
crisandrewsCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

usage

Displays session, daily, and 30-day usage cost summaries in OpenClaw-compatible format.

62|14|Updated Apr 9, 2026
Search Tags:#messaging#monitoring#claude code#tokens#cost tracking#usage#openclaw
One-click install
npx skills add https://github.com/crisandrews/ClawCode --skill usage-crisandrews
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: usage
Source: https://github.com/crisandrews/ClawCode/tree/main/skills/usage
Command: npx skills add https://github.com/crisandrews/ClawCode --skill usage-crisandrews

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Claude Code does not expose session cost data to plugin skills, leaving users without a quick way to check usage costs from messaging channels or the CLI. This Skill provides a standardized cost summary command and transparently falls back to agent resource statistics when real cost data is unavailable. ## Core Features & Use Cases - OpenClaw-Compatible Output: Renders session, today, and last-30-day cost lines in the exact OpenClaw /usage format. - Subcommand Support: Offers /usage tokens, /usage full, and /usage cost variants for compact or verbose displays. - Resource Fallback Reporting: When cost data is unavailable, reports memory directory size, file and chunk counts, and dream event statistics via the agent_status MCP tool and shell commands. - Surface-Aware Formatting: Adapts bold markup for CLI, WhatsApp, and Telegram outputs. - Use Case: A user on WhatsApp asks "cuanto gasté" and receives an honest cost summary noting that native /cost is required for real token data, plus current memory and dream resource stats. ## Quick Start Ask the agent "show my usage cost summary" or invoke /usage to get the session, daily, and 30-day cost report.

Frequently Asked Questions about usage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check Claude Code usage costs from a skill?▼

Invoke /usage to get a cost summary in OpenClaw format. Because Claude Code does not expose session cost data to plugin skills, the output shows n/a for costs and directs you to run the native /cost command in the CLI for real token data.

What subcommands does the /usage command support?▼

The /usage command supports four forms: plain /usage for the default cost summary, /usage tokens for compact token display, /usage full for verbose token and cost output, and /usage cost which matches the default.

Why does the usage skill show n/a for session costs?▼

Claude Code stores session data in internal JSONL files keyed by working directory slug, and token counts are not reliably available to skills. The skill honestly reports n/a and suggests the native /cost command instead of fabricating numbers.

Can I use the usage command from WhatsApp or Telegram?▼

Yes, the skill detects the messaging surface and formats output accordingly, using single asterisk bold for WhatsApp and double asterisk bold for Telegram. It replies through the channel's reply tool when on a messaging surface.

What agent resource statistics does the usage fallback show?▼

When cost data is unavailable, the skill reports memory directory size via du, memory file and chunk counts from the agent_status MCP tool, and dream event counts from the events.jsonl log.

Related Skills

Explore More Software Engineering Skills→
vickysrawatvickysrawat
1

dream-rollback

Reverses a specific Dream consolidation run using the audit trail in memory/dream-log.md.

Community
Intermediate
aws-samplesaws-samples
18

cost-governance

Automate AWS cost tracking, budget alerts, and model downgrade recommendations.

Official
Advanced
tombeliebertombelieber
101

daily-cost

Monitor daily Claude Code spending with aggregate and per-session cost breakdowns.

Community
Intermediate
mferris77mferris77

cost

Calculate token-level costs for sessions and campaigns from Claude Code files.

Community
Advanced