skills.rest
agentic-community
agentic-communityOFFICIAL
Explore all skills by this author →
GitHub

usage-report-v0

Generates LLM-written usage reports from MCP Gateway Registry telemetry exports for comparison against the current deterministic skill.

892|225|Updated May 29, 2025
Search Tags:#comparison#chart generation#Data Analysis#archived#usage report#MCP gateway#Telemetry
One-click install
npx skills add https://github.com/agentic-community/mcp-gateway-registry --skill usage-report-v0
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: usage-report-v0
Source: https://github.com/agentic-community/mcp-gateway-registry/tree/main/.claude/skills/usage-report-v0
Command: npx skills add https://github.com/agentic-community/mcp-gateway-registry --skill usage-report-v0

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires matplotlib, seaborn.

What problem does it solve?

This archived skill preserves the original LLM-driven approach to generating usage reports from MCP Gateway Registry telemetry, enabling side-by-side comparison of prose style, structure, and content against the current deterministic /usage-report skill.

Core Features & Use Cases

  • Comparison-mode reporting: After /usage-report populates a dated folder with telemetry exports, analyzer JSON/CSV files, and PNG charts, this skill writes a v0-style LLM-authored report from the same data.
  • Telemetry analysis workflow: Documents the full pipeline of exporting DocumentDB telemetry via a bastion host, running analyzer scripts, and generating Tufte-style charts (instance distribution, timeseries, lifetime, liveness, LTV spend, install forecast, adoption funnel).
  • Migration auditing: Captures the pre-refactor instructions so teams can verify no semantic content was lost when moving to the template-plus-commentary architecture.
  • Use Case: Run /usage-report for a date, then invoke this skill in comparison mode to produce ai-registry-usage-report-<DATE>-v0.md and evaluate narrative voice differences between LLM-written and deterministic reports.

Quick Start

Run /usage-report first for your target date, then invoke this archived skill in comparison mode to write a v0-style report from the existing data files in the dated folder.

Frequently Asked Questions about usage-report-v0

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

FAQPage Schema
How do I run the archived usage-report-v0 skill in comparison mode?▼

Run /usage-report first for your target date to populate .scratchpad/usage-reports/<DATE>/ with the telemetry export, analyzer outputs, and charts. Then invoke this skill and skip Steps 1-6, starting at Step 7 to write the v0-style report from the existing files.

What is the difference between usage-report-v0 and the current usage-report skill?▼

usage-report-v0 is the archived LLM-driven approach where the model wrote the entire report prose from analyzer outputs, which caused hallucination bugs like incorrect instance counts. The current /usage-report skill uses deterministic template rendering augmented with LLM commentary.

Should I use usage-report-v0 for production usage reports?▼

No. This skill is explicitly archived and marked not meant to be invoked for production reports. Use the current /usage-report skill instead; v0 exists only for side-by-side prose comparison and migration auditing.

What prerequisites does the telemetry export workflow require?▼

You need an SSH key at ~/.ssh/id_ed25519 with bastion access, Terraform state in terraform/telemetry-collector with bastion_enabled set to true, AWS credentials on the bastion host, and an authenticated GitHub CLI for repository stats.

Where are the analyzer and chart generator scripts located?▼

The Python scripts referenced by this skill, such as analyze_telemetry.py, analyze_liveness.py, and the chart generators, live in .claude/skills/usage-report/. This v0 folder contains only the SKILL.md reference document.

Related Skills

Explore More Data & Analytics Skills→
theslashdojotheslashdojo

tunnel

Forward local, remote, and dynamic SSH traffic through bastions.

Official
Advanced
Habitat-ThinkingHabitat-Thinking
44

cost-tracking

Capture AI provider spend and token usage into structured Markdown cost snapshots.

Official
Intermediate
shenmuxingshenmuxing
2

style-review

Review Markdown prose against a 21-rule agent style system.

Community
Advanced
LAF-USLAF-US

session-report

Analyze Claude Code transcripts and render an interactive HTML report.

Official
Intermediate