Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLIAbout this skill
Scaffold a research-grade experiment that tests whether a Looped / Recurrent-Depth Transformer (RDT) is better than a parameter-matched vanilla transformer at vulnerability discovery, when both are continued-pretrained on a security corpus. Use when the user says "test the looped-transformer hypothesis", "run the OpenMythos RDT experiment", "compare loop count to trial budget", "does inference-time loops help security reasoning", or asks for a reproducible research pipeline around the Mythos architecture hypothesis. The skill scaffolds the experiment — it does NOT make claims about Mythos itself, and it explicitly bins OpenMythos as a community speculative reconstruction, not Anthropic's actual model.
Quick start
- Copy the install prompt. Use the green button above this README.
- Send it to your agent. Paste it into your coding agent’s chat and send.
- Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.
Review, license & listing details
This is an imported listing. We haven’t certified its safety, license, or compatibility. Skills may include executable scripts.
Only SKILL.md was scanned. Supporting files and scripts were not checked. No safety grade has been assigned.
0 potential findings · Scanned 11/4/2026 · Scanner 0.2.1
Content SHA-256: 64df2cffc9be5faed5788c527bf424c25a350e6b78d29a9c7460f676efca2a60
About this listing
- Category
- Science & research
- Reported license
- AGPL-3.0
- Listing language (estimated)
- English
- Review status
- Not individually reviewed
- Agent compatibility
- Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.