Skillget

wingman

skill
Guvenlik Denetimi
Uyari
Health Uyari
  • License — License: MIT
  • Description — Repository has a description
  • Active repo — Last push 0 days ago
  • Low visibility — Only 7 GitHub stars
Code Gecti
  • Code scan — Scanned 12 files during light audit, no dangerous patterns found
Permissions Gecti
  • Permissions — No dangerous permissions requested
Purpose
This command-line tool generates visual SVG stat cards and YAML resumes by analyzing your local AI coding assistant usage history. It parses local log and database files from various AI agents (like Claude Code, Gemini, and GitHub Copilot) to calculate statistics.

Security Assessment
Overall Risk: Medium. The primary security consideration is that the tool requires read access to potentially sensitive local files. Specifically, it scans application data directories (such as `~/.claude/`, `~/.local/share/`, and `~/.gemini/`) which contain your AI conversation histories. Exposing or uploading these logs could leak proprietary source code or personal data. The automated code scan found no dangerous code patterns, no hardcoded secrets, and no dangerous permission requests. Additionally, the tool appears to operate entirely offline with no indications of making external network requests.

Quality Assessment
The project is in its early stages, which is reflected by a very low community footprint of only 7 GitHub stars. However, the codebase is actively maintained, with recent pushes, and it uses the standard, permissive MIT license. The automated checks confirm a clean bill of health regarding dangerous coding patterns.

Verdict
Use with caution: The code itself is safe and locally contained, but users should be aware that it reads potentially sensitive conversation logs from their disk.
SUMMARY

Showcase your AI pair usage — SVG cards, resumes, and more.

README.md

Wingman

npm
test
license

Showcase your AI pair usage — SVG cards, resumes, and more.

SVG Card Resume (PDF)
npx @eat-pray-ai/wingman card npx @eat-pray-ai/wingman resume
SVG Card Resume

Supported Agents

Agent Data Source Format
Claude Code ~/.claude/projects/*/*.jsonl JSONL
opencode ~/.local/share/opencode/opencode.db SQLite
Gemini CLI ~/.gemini/tmp/*/chats/session-*.json JSON
Codex ~/.codex/state_5.sqlite SQLite
GitHub Copilot VS Code workspaceStorage/ + globalStorage/state.vscdb JSON + SQLite
MORE Coming soon! TBD

Quick Start

# Generate an SVG stats card (last 90 days)
npx @eat-pray-ai/wingman card

# Generate a rendercv-compatible YAML resume (last 180 days)
npx @eat-pray-ai/wingman resume

Commands

card — SVG Stats Card

# All agents, last 90 days (default)
wingman card

# Specific agents, custom output
wingman card --agents claude-code,opencode -o my-stats.svg

# Date range
wingman card --since 2026-01-01 --until 2026-03-30

# Last 7 days with specific theme
wingman card --days 7 --theme github-dark
Flag Short Default Description
--output -o wingman.svg Output file path
--theme -t github-dark Theme name
--agents all detected Comma-separated agent filter
--since 90 days ago Start date (YYYY-MM-DD)
--until today End date (YYYY-MM-DD)
--days 90 Last N days shorthand
--sections all Comma-separated sections to include

The default github-dark theme renders:

  1. Header — title + date range
  2. Top Stats — tokens (input/output/cache breakdown), estimated cost, sessions
  3. Agent Legend — color-coded bars showing share per agent
  4. Charts — donut chart (token types) + sparkline (daily activity) + model breakdown bars
  5. Activity Heatmap — GitHub-style contribution grid for daily usage
  6. Inventory — plugins, MCP servers, and skills detected across agents
  7. Footer — branding

resume — rendercv YAML Resume

# All agents, last 180 days (default)
wingman resume

# Custom name and headline
wingman resume --name "My Team" --headline "AI Development"

# Specific output path
wingman resume -o my-resume.yaml
Flag Short Default Description
--output -o resume.yaml Output file path
--name Wingman Resume name
--headline AI pair for everything Resume headline
--agents all detected Comma-separated agent filter
--since 180 days ago Start date (YYYY-MM-DD)
--until today End date (YYYY-MM-DD)
--days 180 Last N days shorthand

The generated YAML follows the rendercv schema with sections:

  • Summary — agent count, total tokens, sessions, cost
  • Experience — one entry per agent (sorted by usage), with model breakdowns
  • Education — models grouped by AI lab (Anthropic, Google, OpenAI, etc.)
  • Technologies — plugins, MCP servers, skills inventory

Render your resume at rendercv.com.

How It Works

Agent Adapters → UsageRecord[] → Aggregator → ShowcaseData → Renderer → SVG / YAML
  1. Agent adapters read local data from each AI coding agent (JSONL, SQLite, JSON)
  2. Aggregator groups by agent, calculates totals, builds per-model and daily breakdowns
  3. Pricing engine fetches model costs from models.dev (24h disk cache) to estimate spend
  4. Renderers produce output:
    • Theme renderer → self-contained SVG string (embeddable anywhere)
    • Resume renderer → rendercv-compatible YAML

Development

npm install
npm run dev -- card --days 30        # run directly via tsx
npm run dev -- resume                # generate resume YAML
npm run build                        # bundle to dist/
npx tsc --noEmit                     # type-check
npm test                             # vitest

See AGENTS.md for code style and architecture details.

Extending

Add a new agent adapter

  1. Create src/agents/my-agent.ts implementing AgentAdapter
  2. Register in src/agents/registry.ts

Add a new theme

  1. Create src/themes/my-theme/index.ts implementing ThemeRenderer
  2. Register in src/themes/registry.ts

License

MIT

Yorumlar (0)

Sonuc bulunamadi