Starter. For "can my coding agent run it for me?"
For your AI coding agent
Penthera ships as an Agent Skill, the open format used by Cursor, Claude Code, Claude.ai, and 40+ agent tools. The skill teaches an agent to run authorized scans, pick the right profile, and summarize findings, so you can describe what you want in plain words.
Install (any of these):
# Cross-agent, frictionless
npx skills add danoszz/penthera
# Or the one-line installer (also links the CLI)
curl -fsSL https://raw.githubusercontent.com/danoszz/penthera/main/install.sh | bash
# Or copy into a single project for team sharing
cp -r skills/penthera .cursor/skills/penthera # Cursor
cp -r skills/penthera .claude/skills/penthera # Claude Code
Or as a plugin. The plugin holds only the skill, which runs the pinned CLI; it
starts no MCP server (Anthropic's plugin directory accepts a server that runs
on your computer only from verified publishers). To add the MCP server, run
claude mcp add penthera -- npx -y penthera@2.4.0 mcp (mcp.md).
In Claude Code:
claude plugin marketplace add danoszz/penthera
claude plugin install penthera@penthera
In GitHub Copilot CLI, which reads the same marketplace file:
copilot plugin marketplace add danoszz/penthera
copilot plugin install penthera@penthera
Codex and Gemini CLI take the MCP server instead: see mcp.md.
For Claude.ai (web), zip and upload the skill:
npm run package:skill # produces penthera-skill.zip; upload in Settings > Capabilities > Skills
The skill runs the CLI in your agent's terminal. To give your agent typed, read-only tools instead, add the MCP server: see mcp.md.
Prompts that trigger it:
- "Scan my localhost app on port 3000 for security issues"
- "Run a pre-deploy audit on staging.myapp.com and summarize critical findings"
- "Scan this repo for hardcoded secrets and exposed API routes"
- "Compare today's scan to last week's baseline and show only new findings"
- "Generate a SARIF report for GitHub code scanning"
- "Audit my MCP servers, skills and Claude Code hooks"
The skill adds an authorization gate: it instructs the agent to refuse targets you
do not own or are not authorized to test, and the wizard asks you to confirm before
scanning a remote host. The raw CLI does not check who owns a target. It asks for
confirmation only before active modes (--deep, --fuzz, --adaptive, --all,
--profile deep). You are responsible for every target you pass it. Before the
first scan the agent runs the skill's preflight checks: Node.js 22 or later, a
working penthera command, a reports/ folder, and an authorized target. They
are plain steps in SKILL.md, so the skill runs no script of its own. The Claude
plugin ships Markdown and JSON only, as Anthropic's plugin directory requires. An
install with npx skills add, or a checkout, still has the same checks as a
script, from your project folder:
bash ~/.claude/skills/penthera/scripts/preflight.sh http://localhost:3000
Skill source and workflows live in skills/penthera/SKILL.md.