seanthegeek

Doubleagent — AI skill for Claude Code

AI community

Forensic collection and timeline analysis of AI agent artifacts (Claude Code, Gemini CLI, Antigravity, Codex, Cursor.

How to install Doubleagent

This entry records only its repository, not the path inside it, so there is no exact command to give. Open seanthegeek/doubleagent and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Doubleagent does

Forensic collection and timeline analysis of AI agent artifacts (Claude Code, Gemini CLI, Antigravity, Codex, Cursor, ...) on live hosts and disk images

Alternatives in AI

  • Squeez — Hook-based token compressor for 5 AI CLI hosts (Claude Code, Copilot CLI, OpenCode, Gemini CLI, Codex CLI) 188 ★
  • AI Toolkit — Professional-grade AI coding toolkit: 94 skills, 44 agents, multi-platform (Claude, Cursor, Windsurf, Copilot 167 ★
  • Legal AI Skills — An open collection of 160+ Legal AI Skills for ChatGPT, Claude, Grok, Cursor & Gemini 50 ★

README

doubleagent

Forensic analysis of AI agent use on a system, for incident responders. It covers agents that execute tools or shell commands on a host and leave transcripts, configuration or credentials on disk: Claude Code, Gemini CLI, Antigravity, Codex CLI, Copilot CLI, Cursor, Windsurf, Aider, Ollama, VS Code chat extensions, Hermes, Agent Zero, OpenClaw and others. It has two parts, kept in separate directories because they run in different places and have different dependency rules, and a Docker lab for developers.

Part Directory Runs on
Collectors `collectors/` The host under investigation, or an analyst workstation with a mounted image
Analyzer `analyzer/` The analyst workstation
Agent lab `lab/` A developer's Docker host, to produce real-install fixtures without installing agents

Collectors

Two single-file scripts with the same catalog, manifest schema and archive layout: a POSIX `sh` collector for macOS, Linux and BSD, and a Windows PowerShell 5.1 collector. Each depends on nothing beyond its base system, so it can be uploaded and run once through an EDR remote shell (CrowdStrike RTR, SentinelOne RemoteOps, Defender Live Response, Palo Alto Networks Cortex XDR Live Terminal), run locally by a responder, or pointed at a mounted disk image of any of the three platforms. They also collect Docker and Podman named volumes that belong to known agents. Output is one `tar.gz` (a `.zip` from the Windows collector where `tar.exe` is missing) holding the collected files, a hashed JSONL manifest and a run summary. Host state such as processes and network connections is left to the EDR. With `--inventory` they write nothing and print one JSON line per user and agent found to stdout instead, for a fleet-wide audit through the EDR console.

Usage and options are in [collectors/README.md](collectors/README.md); the list of covered tools, the manifest sc