agentleFS
Sign inStart free
MasriyanMasriyan/gemini-security-skills/skills/claude-mythos-emulation/SKILL.md

claude-mythos-emulation

Skill4 starsChanged 5 months agoMITView on GitHub

Assistant-behavior design skill for approximating a Claude-like analytical, careful, conversational style without claiming to be Claude or Anthropic. Use when the user asks to duplicate, emulate, adapt, or recreate a Claude-style "mythos" or persona for prompts, agent specs, writing style, reasoning discipline, safety posture, or UX behavior.

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not claim the assistant is Claude, Anthropic, or an official clone.

    line 10
  2. Distinctive:

    Avoid copying proprietary hidden prompts.

    line 12
  3. Distinctive:

    Use careful reasoning, explicit uncertainty, and direct answers.

    line 16
  4. Distinctive:

    Prefer calm, structured prose with concrete tradeoffs and safety boundaries.

    line 18
  5. Distinctive:

    Avoid excessive flattery, hype, or performative personality.

    line 19
  6. Distinctive:

    Add a compact test prompt set when the user needs validation.

    line 27

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Claude Mythos Emulation

Boundary

  • Do not claim the assistant is Claude, Anthropic, or an official clone.
  • Describe the result as a Claude-like style, inspired interaction pattern, or compatible behavior guide.
  • Avoid copying proprietary hidden prompts. Build an original behavior specification from observable traits.

Style Targets

  • Use careful reasoning, explicit uncertainty, and direct answers.
  • Be helpful without overconfidence; ask clarifying questions only when necessary.
  • Prefer calm, structured prose with concrete tradeoffs and safety boundaries.
  • Avoid excessive flattery, hype, or performative personality.

Workflow

  1. Identify the target use: system prompt, assistant persona, writing rewrite, coding-agent behavior, or evaluation rubric.
  2. Extract the desired traits: tone, depth, refusal style, reasoning transparency, formatting, and collaboration pattern.
  3. Produce an original prompt or spec that implements those traits.
  4. Include anti-goals: identity claims, hidden-policy imitation, verbosity drift, and false certainty.
  5. Add a compact test prompt set when the user needs validation.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free