elementalsouls

Characterize Agent — Security skill for Claude Code

Security community

Step 1 of an agent pentest — fingerprint the target before sending any payload.

How to install Characterize Agent

Installs to ~/.claude/skills/elementalsouls-claude-agenthunter-characterize-agent/SKILL.md

Terminal

mkdir -p ~/.claude/skills/elementalsouls-claude-agenthunter-characterize-agent && curl -fsSL https://raw.githubusercontent.com/elementalsouls/Claude-AgentHunter/HEAD/commands/characterize-agent.md -o ~/.claude/skills/elementalsouls-claude-agenthunter-characterize-agent/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What Characterize Agent does

name: characterize-agent description: Step 1 of an agent pentest — fingerprint the target before sending any payload. Maps visibility, tool allowlist, egress, oracle semantics, auth/tenancy. Never probe before this. Usage: /characterize-agent

Characterize the target agent the user names. Send NO payload yet — this step decides whether a later probe can even be measured.

Visibility — can you see tool calls / traces, or only final text? (black vs grey box → sets which dete

Alternatives in Security

Memory Self Review — Mine recent agent history (claude-mem + usage stats) for recurring failures and repeated patterns, audit MEMOR 297 ★
Claudit Sec — Security audit tool for Claude Desktop and Claude Code on macOS — single-command visibility into MCP servers 115 ★
SEO God — Turn Claude Code into an agentic SEO operator for your own site — audit, measure, daily loop, AI visibility, s 66 ★